Conflicting Paradigms of Religious Freedom: Liberty Versus Equality Stephen Pepper

BYU Law Review

Volume 1993 | Issue 1 Article 15

3-1-1993 Conflicting Paradigms of Religious Freedom: Liberty Versus Equality Stephen Pepper

Follow this and additional works at: https://digitalcommons.law.byu.edu/lawreview Part of the Constitutional Law Commons, and the Religion Law Commons

Recommended Citation Stephen Pepper, Conflicting Paradigms of Religious Freedom: Liberty Versus Equality, 1993 BYU L. Rev. 7 (1993). Available at: https://digitalcommons.law.byu.edu/lawreview/vol1993/iss1/15

This Article is brought to you for free and open access by the Brigham Young University Law Review at BYU Law Digital Commons. It has been accepted for inclusion in BYU Law Review by an authorized editor of BYU Law Digital Commons. For more information, please contact [email protected]. Conflicting Paradigms of Religious Freedom: Liberty Versus ~~ualit~* Stephen pepper*"

Table of Contents I. INTRODUCTION...... 8 11. BASESFOR CONSTITUTIONALINTERPRETATION ...... 11 A. Text ...... 12 B. History ...... 14 1. Simple History ...... 17 2. The Enlightenment-deist-rationalist View . . . . 18 3. The Radical Protestant View ...... 20 C. Structure and Theory ...... 22 D. Doctrine and Precedent ...... 26 E. Values, Policy and Prudence: The Why" of the Free Exercise Clause ...... 38 F. The Synergism of Text, History, Structure andPolicy ...... 42 111. THE CHOICE-ENLIGHTENMENTVIEW OR RADICAL PROTESTANTVIEW ...... 46 IV. A QUARTETOF DIFFICULTIES...... 49 A. The Establishment Clause, Equality, and Exemptions from Neutral Law ...... 49 B. Drawing Lines, Neutrality and Discrimination . . . 53 C. Administrability and Bureaucracy ...... 56 D. Timidity, Limits and Possibility ...... 59 V. CONCLUSION...... 62

* Based upon presentations at the conference on "New Directions in Religious Liberty" held at Brigham Young University Law School, January 22-23, 1993 and at the Georgetown University Bicentennial Conference, "Religion and the Constitution: Exemptions Based on Conscience," Georgetown University Law Center, April 13, 1989. An earlier version of this paper was published as Stephen Pepper, A Brief for the Free Exercise CZuuse, 7 J.L. & RELIGION323 (1989). ** Professor of Law, University of Denver College of Law. A.B. 1969, Stanford University; J.D. 1973, Yale Law School. 8 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [I993

Should the Free Exercise Clause of the First Amendment be read to shelter religiously motivated conduct from otherwise valid, generally applicable laws? There is general agreement, both on the Supreme Court and among commentators, that laws aimed at religion-laws intentionally singling out a religious practice, or religion in general, for a burden or prohibition-are presumptively invalid under the First Amendment.' But what of laws which are not designed to affect religion; laws with recognized, legitimate governmental purposes which have the added consequence of impinging upon, burdening, or prohibiting religious conduct? Four examples from well-known cases illustrate the issue. First, consider Mrs. Frances Quaring, a Nebraska housewife who believed that the Second Commandment prohibited her from having her photograph taken or from carrying or using photographs ("graven images" from her perspective). The State of Nebraska, however, required that applicants for a driver's license have their photographs taken and affixed to the license and prohibited driving without a valid driver's license. Thus, by state law Mrs. Quaring was denied that primary instrument of freedom and mobility in modem America-use of the automobile-because of her religious beliefs. Should we understand the Constitution's guaranty of freedom of religion as protecting Mrs. Quaring's mobility?2 Next, consider the situations in which activities of specific religions have been criminally sanctioned. The Amish are bound by a religious command to "live separate and apart"

1. Curiously, under modern doctrine it is the Establishment Clause which has been held to prohibit gover~lental action which intentionally or explicitly discriminates against religion or which discriminates among religions. See Larson v. Valente, 456 US. 228 (1982); McDaniel v. Paty, 435 U.S. 618, 636-42 (1978) (Brennan, J., concurring in the judgment); Everson v. Board of Educ., 330 U.S. 1, 15-16 (1947). With the decision this term in the Santeria animal sacrifice case, the Court has held that the Free Exercise Clause also prohibits laws which discriminate against religious conduct. Church of the Lukumi Babalu Aye v. City of Hialeah, 113 S. Ct. 2217 (1993). 2. When the Supreme Court considered this case, it was unable to answer that question. The Eighth Circuit held that the state must issue Mrs. Quaring a license without a photo. Quaring v. Peterson, 728 F.2d 1121 (8th Cir. 1984), afd by equally divided Court sub nom. Jensen v. Quaring, 472 US. 478 (1985) (per curiam, Powell, J., did not participate). 71 CONFLICTING PARADIGMS 9 from mainstream society. They understand this as a command from God which requires that their children be raised "separate and apart." In turn, to the extent formal education is allowable at all, it must take place in small local schools and may not go beyond the eighth grade. Most states, however, have promulgated laws mandating universal education of all children up to the tenth grade. These laws were not targeted at the Amish, but they render criminal the religiously mandated conduct of the Ami~h.~ At one time in the history of the Mormon Church, polygamy was a religious duty for those members able to practice it. However, being married to more than one wife at a time was also the traditional crime of big am^.^ For many American Indians, the use of peyote is the central act in the religious practice of the Native American Church. However, in many states, use of peyote (along with many other powerful hallucinogenic drugs) is criminal under generally applicable, formally neutral laws.5 Except in the case of the Mormons, the government did not intentionally discriminate against or

3. In Wisconsin v. Yoder, 406 U.S. 205 (1972), Amish parents who refused to send their fourteen- and fifteen-year-old children to school after the eighth grade were convicted of violating the state's compulsory school attendance law. The Supreme Court held that the Free Exercise Clause sheltered the Amish from criminal prosecution for this conduct. The Court has also held, to the contrary, that the Clause does not shelter the Amish in their belief that the mandate to live "separate and apart" requires that they neither contribute to nor receive benefits , from the Social Security system. United States v. Lee, 455 U.S. 252 (1982). Yoder is examined at some length in Stephen Pepper, Reynolds, Yoder, and Beyond: Alternatives for the Free Exercise Clause, 1981 UTAH L. REV. 309, 333-45 [hereinafter Pepper, Alternutives]; and I have discussed Lee in Stephen Pepper, The Conundrum of the Free Exercise Clause-Some Reflections on Recent Cases, 9 N. KY. L. REV. 265, 299-302 (1982) [hereinafter Pepper, Conundrum]. These cases are also considered briefly infia part 1I.D. 4. In the initial Supreme Court case dealing with the Free Exercise Clause, the bigamy conviction of a Mormon was upheld. Reynolds v. United States, 98 U.S. 145 (1878). The opinion is described briefly infia part ED., and examined in some detail in Pepper, AZternutives, supra note 3, at 317-26. Although neutral in form, the legislation enforced in Reynolds may in fact have been targeted specifically at the Mormons. Orma Linford, The Mormons and the Law: The Polygamy Cases, 9 UTAHL. REV. 308, 314-19 (1964). If so, it would now be considered presumptively invalid, and would be factually somewhat similar to the Hialeah case. See supra note 1. 5. Sacramental usage of peyote by members of the Native American Church was held unprotected by the First Amendment in Employment Division v. Smith, 494 U.S. 872 (1990). The Supreme Court's opinion is examined at length in Douglas Laycock, Th Remnants of Free Exercise, 1990 SUP. CT. REV. 1; Michael W. McConnell, Free Exercise Revisionism and the Smith Decision, 57 U. CHI. L. REV. 1109 (1990). 10 BRIGHAM YOUNG UNIWRSITY LAW REVIEW [I993 burden these groups in their religious conduct; yet their members faced criminal conviction for following the tenets of their religion. In these situations religiously motivated conduct is either penalized or prohibited by laws which are not aimed at religion. Although the law in each case is religiously neutral in form and intent, religious believers may be burdened by the law's mandate in a way quite different from that law's effect on the rest of society. Should such situations be governed by a constitutional paradigm of equality, or a paradigm of liberty? William Marshall, a friend and a colleague in this symposium, argues that equality is the proper paradigm, and that formal equality, rather than substantive equality, should regulate.' According to this view, it is unfair-and contrary to the underlying logic of the First Amendment-to prefer religion as a basis for action over other beliefs that ground one's conduct. If the First Amendment prevents criminalization of the conduct only if it has a religious motivation, but allows governmentally imposed punishment or burden if the same conduct is otherwise motivated, religion has been preferred over other bases for action; and such a result is not constitutionally acceptable. The wrong or unfair aspect of this unequal treatment may be trivial in some instances and substantial in others. Aside from religious reasons such as Frances Quaring's, it is hard to imagine substantial prejudice or harm in being required to have a photo on one's driver's license. It simply is not much of a burden. Peyote usage is usually a rather unpleasant and -cult experience. Although there are certainly some who wish to try it on the basis of non-religious reasons, either as a result of intellectual interest or a sense of adventure, the burden of being prohibited from doing so does not seem great.

6. William P. Marshall, The Case Against the Constitutionally Compelled Free Exercise Exemption, 40 CASE W. RES. L. REV. 357 (1989-90) [hereinafter Marshall, The Case Against the Compelled Exemption]; William P. Marshall, Solving the Free Exercise Dilemma: Free Exercise as Free Expression, 67 MINN. L. REV. 545 (1983) [hereinafber Marshall, Free Exercise as Free Expression]; see also Ellis West, The Case Against a Right to Religion-Based Exemptions, 4 N~REDAME J.L. ETHICS & PUB. PoL'Y 591 (1990). A notion of substantive, as opposed to formal, equality may well lead to results quite different than those suggested by Marshall and West. See infia part IVA. Professor Marshall's contribution to this symposium discusses a different problem, although from a related perspective. William P. Marshall, The Inequality of Anti-establishment, 1993 B.Y.U. L. REV. 63. 71 CONFLICTING PARADIGMS 11 On the other hand, there are many people with substantial, strongly held, non-religious reasons for wanting to educate their children away from public or governmentally regulated schools. To allow only those with religious reasons to avoid the compulsory attendance law is a burden of a different order from the first two examples. Similarly, there are many with strong reasons for wanting a form of marriage not sanctioned by the state. Homosexual couples with long term marriage-like relationships are one example. To allow those with sincere religious reasons to avoid the laws concerning marriage while they remain applicable to all others appears to be a substantial wrong. The First Amendment contains a guaranty of the "free exercise" of religion.' Is religion singled out in the Amendment for special treatment or for equal treatment? Is freedom for the religious dissenter the essential purpose of the Free Exercise Clause, or is it equality between those acting upon religious motives and those acting upon other bases? In elaborating an answer to these questions, the discussion in Part I1 will canvass the major categories of constitutional discourse as bases for interpreting the Free Exercise Clause and provide a response to some of the related arguments of Professor Marshall. Parts I11 and IV will then turn to the basic interpretive choice presented by the issue of religiously based exemptions from neutral law, and to a few of the difficulties that must be faced in developing a vigorous free exercise doctrine based upon religious liberty.

11. BASESFOR CONSTITUTIONALINTERPRETATION Constitutional interpretation occurs, for the most part, through argument which can be divided into five major categories: (1) text, (2) history, (3) structure and theory, (4) doctrine and precedent, and (5) prudence, social policy and justi~e.~These categories are interrelated, lack clear boundaries, and can each be subdivided into narrower aspects. The fifth category, in particular, is hard to define, capacious, and capable of being divided into numerous subcategories. These categories do provide, however, a useful path for analysis

7. U.S. CON^. amend. I. 8. PHILIP BOBBITT,CON~IONAL FATE, 3-119 (1982); Richard H. Fallon, Jr., A Constructivist Coherence Theory of Constitutional Interpretation, 100 mxJ.L. REV. 1189, 1189-209 (1987). 12 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [I993

of the meaning of a constitutional provision. A. Text The words of the First Amendment clearly single out religion for both a benefit and a burden. "Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof. . . ."g On its face, the language grants a unique advantage to religious conduct, protecting it from governmental imposition; and imposes a unique disadvantage, preventing the government from supporting it. To understand this as a provision which puts religion on an equal footing with other bases for action seems to be a curious reading. There are no "free exercise" or "establishment" provisions for science, sports, philosophy or family relations. The language itself thus seems to answer whether we have a paradigm of equality or of liberty; the language of the Free Exercise Clause is clearly in the form of a grant of liberty.'' It must also be noted that this grant of liberty is in language as absolute as that found anywhere in the Constitution. Other provisions in the Bill of Rights contain moderating quali- fiers: "unreasonable searches and seizures,"" c'excessiue bail,"12 "cruel and unusual p~nishments,"'~"due process of law."'* The words of the Free Exercise Clause, by contrast, seem pure and inclusive: no law prohibiting the free exercise of religion.'' Only the protections of speech and press are as ab-

9. U.S. CONST.amend. I. 10. Laycock, supra note 5, at 13-14. 11. U.S. CONST.amend. IV (emphasis added). 12. Id. amend. VIII (emphasis added). 13. Id. (emphasis added). 14. Id. amends. V & XIV (emphasis added). 15. One could, of course, focus on words other than those emphasized. See Pepper, AZtenatives, supra note 3, at 353 11.195. For example, "prohibit" is a different usage than "abridge," which is the usage applied to other First Amendment freedoms. Does this mean that government can impinge on religious conduct short of outright prohibition, while it cannot even abridge speech and press? Government could prohibit Jews or Catholics from voting or holding office, or tax them at a higher rate, under such a view. But those seem exactly the kind of abuses that the framers intended to prevent, and I know of no scholar or court which has seriously suggested such a wooden reading. Or one could focus on "Congress" as the only governmental entity limited by the Amendment, thus leaving all other governmental actors free to impbge on religious conduct as they see fit. But the federal government ads, for the most part, through delegation of congressional authority. (Thus, the IRS would clearly be covered.) And there has been no inclina- tion that I know of to interpret the First Amendment differentidy in its application to the judiciary and executive, as the "Congress" usage might suggest. More- 71 CONFLICTING PARADIGMS 13 solute in terminology. Speech and press are inherently limited in their impact, however, because they are a particular, limited species of conduct. They refer to communication only, and as such their consequences are inherently limited: they have an impact directly only on other human minds; they can cause no direct injuries because further consequences must be mediated through the recipient of the communication. "Exercise," to the contrary, denotes action in general, and "religion" is not limited in its ordinary understanding to communicative acts. "Exercise of religion" covers religiously mandated smuggling and shelter of aliens (the "sanctuary" movement), smuggling of slaves from slave states to free (the underground railroad), human sacrifice, the sacramental drinking of wine, refusal to send one's children to public school, and so ,on. Neither "religion" nor "exercise" provides the kind of inherent limits that come with "speech" and "press." This distinction is demonstrated within the First Amendment itself. Assembly . . . "to petition the Gov- ernment for a redress of grievances" is included in First Amendment protection,16 and it is communication. But unlike speech and press, "assembly" is not inherently limited, and thus the First Amendment protection is limited in its terms to "peaceabl[ey assembly only." No similar modifier appears in regard to exercise of religion, thus making this freedom the most absolute in its terminology of any in the Constitution. Perhaps the absence of a modifier is an accident and without meaning. But the framers had more limited models at hand, which they did not choose. An examination of language available to, but not chosen by, the framers crosses into an area of both history and text. During the same period when it was drafting the First Amendment, Congress adopted the 1787 Ordinance for the Government of the Northwest Territory, providing that: "No person, demeaning himself in a peaceable

over, the other nine amendments in the Bill of Rights do not share this usage, suggesting it had no limiting meaning. The religion clauses certainly were not intended to apply to the states, but the application of the Bill of Rights to the states is a large and complex issue. See, e.g., MARKD. HOWE,THE GARDENAND THE WILDERNESS19-23 (1965); WILBERG. KATZ, RELIGIONAND AMERICANCONSTI- TUTIONS 8-10 (1964). Suffice it to note that if the other provisions of the Bill of Rights are held applicable to the states, there is no reason not to also apply the Free Exercise Clause. 16. U.S. CONST.amend. I. 17. Id. 14 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [ 1993 and orderly manner, shall ever be molested on account of his mode of worship, or religious sentiments, in the said territory."18 Similarly, Jefferson's Bill for Establishing Religious Freedom, introduced in Virginia in 1779, passed in 1785, and well known by the framers of the First Amendment, provided: [Nlo man. . . shall be enforced, restrained, molested, or burthened in his body or goods, nor shall otherwise suffer, on account of his religious opinions or belief; but that all men shall be fiee to profess, and by argument to maintain their opinions in matters of religion, and that the same shall in no wise diminish, enlarge, or effect their civil capacities.lg The absoluteness of the language of the Free Exercise Clause, so different from both of these then-available models concerning freedom of religion and from the phrasing of other civil liberties in the Constitution, is consistent with an earlier artic- ulation of James Madison in a seminal document of American religious freedom, his A Memorial and Remonstrance: 'We maintain therefore that in matters of Religion, no mans [sic] right is abridged by the institution of .Civil Society, and that Religion is wholly exempt from its cognizance."20Significantly, Madison drafted the original version of the First Amendment and was the moving force in Congress behind its passage.21 B. History The history surrounding the religion clauses is a large subject, and one much written about. Most of that writing has tended to focus upon the Establishment Clause," although it

18. 1 ANSON P. STOKES, CHURCHAND STATEIN THE UNITEDSTATES 480 (1950) (emphasis added). 19. Id. (emphasis added). 20. JAMESMADISON, A MEMORIALAND REMONSTRANCE(1785), reprinted in JAMESMADISON ON RELIGIOUSLIBERTY 55, 56 (Robert S. AUey ed., 1985). 21. WI~L. MILLER, THE FIRST LIBERTY120-21 (1986); STOKES, supra note 18, at 339-50, 538-48. It is worth noting that Madison used the phrase "free exercise of Religion" as early as 1776 in his suggested language for the Virginia Decla- ration of Rights, Hunt, James Madison and Religious Liberty, Am. Hist. A. AM. Rep. for the Year 1901 at 166-67 (1902), and that he used the same language from the Declaration in his 1785 Remonstrance. 22. A notable recent exception is Michael W. McCo~ell,The Origins and His- torical Understanding of Free Exercise of Religion, 103 HARV. L. REV. 1409 (1990), which provides a thorough examination of the subject. What follows in this brief section addressing the history of the clauses is in accord with the evidence presented by Professor McCo~ell,and is consistent with his account, although it was first written prior to publication of his work. See supra note *. 71 CONFLICTING PARADIGMS 15 is the same history behind both clauses. Because it is a legal document we are attempting to interpret, history as it relates to the intentions and meaning of those who framed and adopted the First Amendment is surely relevant. Because it is a constitution we are interpreting, a document intended to be lasting and therefore at least somewhat elastic and vague, that relevant history may well not be determinative. Here, &er a few preliminary observations, I will elaborate on three aspects of the historical background which are helpful in contemporary interpretation of the Free Exercise Clause and point toward the conclusion that the relatively absolute language of the Free Exercise Clause may not have been an accident. As I have suggested elsewhere, the clearest, least controversial, narrow understanding of the original purpose of the religion clauses is to see them as aimed at two different aspects of the perceived evil of government affiliation with religion.23 The Establishment Clause was aimed at prohibiting federal government support for one or several sects through affirmative provisions such as subsidies or declarations of dogma. The Free Exercise Clause was aimed at prohibiting support for one or several favored sects through negative provisions aimed at other disfavored sects such as limitation of the franchise, im- prisonment, or banishment. Certain provisions of religious "establishments" with which the colonists were familiar could be categorized under either heading; for example, legally mandated church attendance. It is accurate to characterize this intention as both "disestablishment" and as "separation of church and state."24 Moving beyond this narrow understanding to resolve contemporary issues on the basis of history is very diffcult because those who supported this "disestablishment" and "separation of church and state" did so for very different motives.25 There were at least three significant streams of thought coming together to support adoption of the religion clauses, two of which will be discussed further below. The Enlightenment- deist-rationalist view perceived separation primarily as protect-

23. See Pepper, Altenatives, supra note 3, at 313. 24. 'Disestablishment" is somewhat inappropriate because there was no federal establishment to begin with. "Separation of church and state" is somewhat inappropriate because it has a potential meaning far larger than the narrow meaning given above. 25. Pepper, &ternatives, supra note 3, at 311-17. 16 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [I993 ing the state-this new experimental federal governmenefrom the likely disruptive and possibly destructive influence of organized religion. The radical protestant view, to the contrary, saw separation as protection for the church from the profane and corrupt influence of the state. In number, these two streams were far from a majority of those supporting separation, but they had influence far beyond those numbers. The Enlighten- ment-deist-rationalist stream, exemplified by Jefferson and Madison, was prominently represented among the educated elite and was the preeminent influence in constructing the theory and drafting the text of the Constitution. It is thus no accident, although it is certainly surprising, that the Constitu- tion is a so thoroughly secular document, with but three pass- ing references to anythmg religious. The radical protestants, on the other hand, were the "outs," whose church was not estab- lished both because they were the minority and because they believed in separation as a religious principle. To a significant extent and particularly in Virginia, they provided the political steam to power the movement for disestablishment and separa- tioaZ6 The third and probably largest stream of thought perceived the religion clauses of the First Amendment as a federalism provision. These were the "ins." Content with the religious establishments that were currently the status quo in their home states, disinclined to a contest with other religions over support from the new federal government, and somewhat .anx- ious about who might win if the federal government were allowed to support and inhibit religion in the way the colonial governments had in the recent past, this group supported the Establishment and Free Exercise Clauses as limits on the federal government only, leaving the states free in the realm of church and state." Given the vast growth in federal government, this view would leave substantial coverage for the kee Exercise Clause: the Bob Jones casez8involved federal law as did recent cases concerning social security numbers:' use of

26. RICHARDE. MORGAN,THE SUPREMECOURT AND RELIGION4-26 (1972); HOWE,supra note 15, at 1-31; MILLER,supra note 21, at 120-21. 27. See HOWE,supra note 15, at 19-23; KATZ, supra note 15, at 8-10. 28. Bob Jones Univ. v. United States, 461 U.S. 574 (1983); see i@a discussion accompanying notes 84-87. 29. Bowen v. Roy, 476 U.S. 693 (1986). 71 CONFLICTING PARADIGMS 17 national forest lands:' and yarmulkes in the military.31 But most cases concern state law, such as Nebraska's driver's license rules and Oregon's criminal law concerning usage of peyote, and here the clauses would not apply. The federalism view has become an anachronism. Over the last hundred years the Supreme Court has incorporated most of the provisions of the Bill of Rights generally, and the First Amendment in particular, into the Due Process Clause of the Fourteenth Amendment and applied them to the states. Wheth- er or not this was justified in terms of the intentions and purposes of those who adopted the Fourteenth Amendment, a very large and controversial topic, it is now quite clearly the law of the Constitution-a de facto binding constitutional amendment. Thus the still relevant historical purposes and meanings are primarily those of the Enlightenment-deist-rationalistview and the radical protestant view. Before turning to these perspectives, however, some basic historical context is useful.

I. Simple history It is important to remember that the notion of a secular government was a new and relatively untried idea at the time of the American revolution. The relation between church and state was not a marginal problem for the colonists as it is for us. Civil strife over issues of religion and state was a more or less constant problem in the historical background of the colonists who were engaged in creating new independent states and a new federal government when the religion clauses were adopted. Not far in the past was a history of armed conflict and bloodshed over issues of religion and state. The long-term background was the Reformation and the several hundred years of armed European conflict over state affiliation with one religion or another. The more recent history for most was the less bloody English background of constant strife and suspicion: first, the break from Roman Catholicism under Henry VIII; next, the attendant fear of both external and internal Catholi- cism; and finally, the Anglican-Puritan conflict. Many of the colonies had been created by groups seeking shelter from these conflicts and the burdens of practicing a religion disapproved by the state. And because the alliance of church and state was

30. Lyng v. Northwest Indian Cemetery Protective Ass'n, 485 U.S. 439 (1988). 31. Goldman v. Weinberger, 475 U.S. 503 (1986). 18 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [I993 the only available model, that alliance, and its attendant strife, was then replicated in the colonies. These were not distant problems. In 1774 Madison wrote to a friend about "five or six well-meaning men in close jail [in the adjacent county] for publishing their religious sentiments, which in the main are very or tho do^.'"^ Being put in jail for your relatively "orthodox" religious opinions can make you rather angry-probably fighting angry. That, and far worse, was the reality of the history of the alli- ances between church and state for those who were drafting and adopting the First Amendment; that was the patrimony and the dilemma facing those who were creating a new government. As a matter of historical fact the alliance of church and state meant trouble. It was that problem, well-known and general, that the framers of the First Amendment were trying to ameliorate with their experiment in the separation of church and state. Part I1.E. considers more directly what that suggests for contemporary interpretation of the Free Exercise Clause.

2. The Enlightenment-deist-rationalist view Consistent with the clearest narrow understanding of the historical intention of the religion clauses noted above, it is common now to assert that both clauses had a single primary purpose: freedom of religion.33 The free exercise provision directly protected that freedom while the non-establishment provision was perceived as the primary necessary instrumental provision to assure it indirectly. While there is probably some truth to this assertion, it leaves out a major concern of those approaching the issue fiom the Enlightenment perspective. From that approach, the alliance of church and state threatens good government in general, not just in relation to freedom of religion." From this perspective, the lesson of history was that the alliance tends toward absolutism in government and toward political division and strife over religious issues. Free- dom in general (limited government) and freedom of inquiry in particular were crucial to progress and good government. A church formally co~ectedto the state was seen as likely to be hostile to such freedom. In this view, organized religion is per-

32. LEO PFEFFER, CHURCH, STATE,AND FREEDOM91 (rev. ed. 1967). 33. See, e.g., LAWRENCEH. TRIBE,AMERICAN CONSTITUTIONAL LAW 1160-61, 1201-04 (2d ed. 1988). 34. MORGAN,supra note 26, at 16-18. 71 CONFLICTING PARADIGMS 19 ceived as backward and threatening. It is impossible to read Jefferson's Bill for Establishing Re1 igious Freedom or Madison's A Memorial and Remonstrance, for example, without noticing this perspective. Madison, in one of many examples from the Remonstrance: What influence in fact have ecclesiastical establishments had on Civil Society? In some instances they have been seen to erect a spiritual tyranny on the ruins of the Civil authority; in many instances they have been seen upholding the thrones of political tyranny: in no instance have they been seen the guardians of the liberties of the people.35 This view is an inversion of the commonly stated one above that both clauses have freedom of religion as their primary purpose. Here, to the contrary, the Establishment Clause is seen as crucial to good government, its purpose is good government in general;36and the Free Exercise Clause is a necessary instrument to support the primary Establishment Clause goal of separating the churches from the government. From this perspective freedom of religion also may have been seen as a significant good in itself, but that was of secondary importance. 37

35. MADISON,supra note 20, at 58. 36. This understanding of the Clause continues to have substantial vitality. The mixing of government and religion can be a threat to free government, even if no one is forced to participate. When the government puts its imprimatur on a particular religion, it conveys a message of exclusion to all those who do not adhere to the favored beliefs. A government cannot be premised on the belief that all persons are created equal when it asserts that God prefers some. Only "[alnguish [sic], hardship and bitter strife" result "when zealous religious groups struggl[e] with one another to obtain the Government's stamp of approval." Such a struggle can "strain a political system to the breaking point." Lee v. Weisman, 112 S. Ct. 2649, 2665-66 (1992) (Blackmun, J., concurring) (quoting Engel v. Vitale, 370 U.S. 421, 429 (1962) and Walz v. Tax Comm'n, 397 U.S. 664, 694 (1970) (opinion of Harlan, J.) (citations & footnote omitted)). 37. Steven Smith denominates this the "civil peace rationale" for religious freedom and acknowledges that it "probably influenced the founding generation." Ste- ven D. Smith, The Rise and Fall of Religious Freedom in Constitutional Discourse, 140 U. PA. L. REV. 149, 163-66 (1991). He argues, however, that to consider it the "primary rationale" is probably "anachronistic," pointing out that even Jefferson and Madison relied substantially on other reasons. Both Smith and McComell have concluded that the radical protestant view, to be discussed infra, was more common and influential. Id.; McCo~ell,supra note 22. This is probably true, but Smith probably deemphasizes the Enlightenment view too much. Particularly in light of the general prevalence of a religious point of view, documented by Smith and McConnell among many others, the Constitution's lack of reference to God and 20 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [I993

There is a sub-element of this view which should be mentioned briefly here, and will be returned to later. Enlighten- ment thought is rational, skeptical and scientific. It is both a symptom and a cause of secularization. The American Constitu- tion, a strikingly secular document, is clearly the product of this thought. The Establishment Clause makes sense in this scheme as a guarantor of secular government. But the Free Exercise Clause is harder to understand. I have suggested in previous writing that the Free Exercise Clause may well repre- sent the compromise that those supporting the relatively new secular world view were willing to make with the preexisting, and hitherto dominant, religious world view.38 The bounds of that compromise do not of themselves appear from the language of the clauses or this history. And had the issue been addressed directly, there would probably not have been agreement even among those working from the Enlightenment view,39 not to mention between those who shared this view and those who remained firmly attached to a religious view. This notion of a compromise between those of the Enlighten- ment-deist-rationalist view and those with a strong and widely held religious view makes more sensible the surprisingly absolute language of the Free Exercise Clause and may suggest that real power was contemplated for the Clause even by those primarily in the Enlightenment stream of thought. In this regard there is some striking social contract imagery used by Madison that will be quoted in the Structure and Theory section below.

3. The radical protestant view The radical protestant view shared with the Enlighten- ment view a firm belief in separating church and state, but the

religion, see infra note 42, is both surprising and significant. And Jefferson and Madison, who resonated with and articulated the Enlightenment view quite clearly, were pivotal and influential adors in the history of religious liberty and the construction of the First Amendment. Each will be discussed briefly as the discussion proceeds below. As to Madison, see supra note 21; MADISON,supm note 20; and Stephen Pepper, Taking the Free Exercise Clause Seriously, 1986 B.Y.U. L. REV. 299, 301-03 [hereinafter Pepper, Taking]; see also supra part 1I.C. As to Jefferson, see Pepper, Alternatiues, supra note 3, at 319-20, and sources cited therein. 38. See Pepper, ALternatiues, supra note 3, at 378; Pepper, Taking, supm note 37, at 304-06. 39. A clear difference between Jefferson and Madison will be mentioned at the beginning of the Doctrine and Precedent discussion, infra part 1I.D. 71 CONFLICTING PARADIGMS 21 motive was quite different: it was the church which was threat- ened by and needed to be removed from the state, not the state which needed protection. And religion needed protection from the state's support as well as its potential hostility; official governmental support was a corruption to the purity of the church. Thus from the believer's side it is the Free Exercise Clause which is of most importance-a guaranty that the church will not be interfered with by government-and the Establishment Clause takes on the more instrumental role. Roger Williams was the initial and eloquent proponent of separation as a necessity for religion, and his image of the "garden" of religion needing protection from the "wilderness" of the secular world, including the state, is e~emplary.~'The community apart, the chosen, the need for separation from the errant world, is a recurrent and strong theme in protestantism. Cre- ation of a community of truth requires shelter from the larger community, including its perhaps well-intentioned laws. A direct line runs from this understanding to the situation of the Amish, who refuse to send their children to public school or to participate in the social security system, and to Mrs. Quaring. The intervening centuries change the perspective of the insular religious community little, but they add a pervasiveness of law (public schools, driver's licenses, a voluminous ubiquitous tax code) that makes separation of garden from wilderness a far more daunting task. In his foreword on the Bob Jones case, Robert Cover made an observation which exemplifies this view and which resonates with the compromise that the Enlightenment understanding may have been making with the adoption of the Free Exercise Clause: There is a powerful, almost physical image at work in the conception to which the Amish and Mennonites implicitly appeal in their constitutional confession. The image is one of a dedicated, sacred space, a refuge carved out from the general secular, legal space of the state. Within the dedicated nom- ic refuge, there is an accommodation to a religious rule of recognition expressed in Acts 529-"We ought to obey God rather than men"-instead of submission to the principle, embodied in article VI, section 2 of the Constitution, that "[tlhis Constitution, and the Laws of the United States which shall be made in Pursuance thereof. . . shall be the supreme

40. See generally HOWE,supra note 15, at 1-31. 22 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [I993

Law of the land

C. Structure and Theory As noted above, the Constitution is a remarkably secular document creating a secular government." The Establishment Clause appears thoroughly consistent with that document in that it attempts to assure that the secular government remains secular. The Free Exercise Clause, to the contrary, contains a remarkably absolute protection for religious conduct dacult to square, on its face, with the rest of the Constitution. From the radical protestant perspective, of course, this absolute protection is perfectly sensible: the government is left free to be what the government wants to be, but space must be left for the church to be what it must be. The Free Exercise Clause creates that space. From the Enlightenment side the apparent capa- ciousness of that space is more difficult to explain. James Mad- ison, using the imagery of the social contract, may supply the answer. He defines religion as "the duty which we owe to our Creator and the manner of discharging it,"'3 and then asserts: This duty is precedent, both in order of time and in degree of obligation, to the claims of Civil Society. Before any man can be considered as a member of Civil Society, he must be considered as a subject of the Governour of the Universe: And if a member of Civil Society, who enters into any subordinate Association, must always do it with a reservation of his duty to the General Authority; much more must every man who becomes a member of any particular Civil Society, do it with a saving of his allegiance to the Universal Sovereign.44

41. Robert M. Cover, The Supreme Court, 1982 Term--Foreword: Nomos and Narrative, 97 HARV. L. REV. 4, 30 (1983). 42. The text of the Constitution has only three references to religion; two are oblique and the third excludes religion from political relevance: (1) the clause exempting Sundays as days to be counted in determining the period of time within which the President must exercise his veto; (2) the dating of the document as "in the year of our Lord one thousand seven hundred and eighty seven," and (3) the crucial clause of Article VI [proscribing] religious tests for office. MORGAN,supra note 26, at 20. 43. MADISON,supra note 20, at 56 (quoting VA. DECLARATIONOF RIGHTSart. XVI (1776)). 44. Id. 71 CONFLICTING PARADIGMS 23

This passage proceeds to the conclusion quoted above45that religion is "wholly exempt" from the rule of government, a conclusion consistent with the absolute language of the Free Exercise Clause. The procedural nature of the Constitution as a whole is congruent with this view of religion as a subject set aside from the general jurisdiction of government. The Constitution creates structures and processes for making binding communal decisions; that is, it creates governmental mechanisms for making and implementing law.46 The Constitution does not an- nounce ultimate truths and it does not create mechanisms for discovering or announcing ultimate truths. (Note that it is the Declaration of Independence that announces 'We hold these truths to be self-evident . . . The Constitution has other business at hand.) Ultimate truths or reality are left in the non-governmental sphere; they are for individuals or groups to discover. Truth as seen by the individual or group may form the basis for their actions within the processes of government? it may provide the motive for law making and application; but truth is neither the object nor the result of governmental processes. This view of the Constitution provides an integral understanding of the First Amendment under which at least one of its purposes is to keep ultimate truth, and the mechanisms for finding it, outside of the sphere of government control. The connection between the freedoms of speech and press and the truth are well articulated in constitutional law. The place of religion beside them in the First Amendment is understand- able: in the view of the believer it represents ultimate truth, the ground from which all else proceeds. The Establishment Clause prevents ultimate truth (or those who believe they know it) from impinging too directly upon both government and those who may not share that understanding of the truth. The Free

45. See supra text accompanying note 20. 46. The almost exclusively process oriented nature of the Constitution is one of the primary points of John H. Ely's DEMOCRACYAND DISTRUST(1980). The choice among processes of course involves substance. See Paul Brest, The Substance of Process, 42 OHIO STATEL.J. 131 (1981). 47. THE DECLARATIONOF INDEPENDENCE para. 2 (US. 1776). 48. See generally Frederick M. Gedicks, Some Political Implications of Religious Belief, 4 NOTRE DAME J.L. ETHICS & PUB. POL? 419 (1990); Frederick M. Gedicks & Roger Hendrix, Democracy, Autonomy and Values: Some Thoughts on Religion and Law in Modern America, 60 S. CAI,. L. REV. 1579 (1987). 24 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [I993

Exercise Clause prevents the government from impinging upon the ultimate truth of the belie~er.~' From a different structural perspective, the religion clauses are prominent as one of very few provisions in the Constitution which clearly support mediating institutions. Religion, as understood both by the framers and contemporary observers, is primarily a group phenomenon; it is very diEcult to read the clauses without perceiving an intention to create a space apart for religion, and therefore for religious institution^.^' The clauses are "liberal" in the sense of focusing upon liberty and restraining government, but they are more "civic republican" and communitarian in their support for one of the primary intermediate institution^.^^ The Free Exercise Clause may well be the only explicit support for non-governmental community in the Con~titution,~~and this is of major significance. We have here no image of the "self-constituted" individual, no "liberalism" of lonely, isolated individualism." Rather we have a space, a liberty, for groups and for individuals. And given the prominence of the radical protestant vision, there is reason to think of the group as constituting the individuals to as significant an extent as the individuals constituting the group* Thus we have three different but complementary theoretical perspectives on the Free Exercise Clause. It can be seen as a Lockean social contract provision protecting liberty and re-

49. See generally Nomi M. Stolzenberg, "He Drew a Circle That Shut Me Out": Assimilation, Indoctrination, and the Paradox of a Liberal Education, 106 HARV. L. REV. 581 (1993). The right to assemble to petition for redress of grievances is the .one First Amendment right that does not fit easily into this understanding. 50. "The religion clauses of the Constitution seem to me unique in the clarity with which they presuppose a collective, norm-generating community whose status as a community and whose relationship with the individuals subject to its norms are entitled to constitutional recognition and protection." Cover, supra note 41, at 32 n.94. See Frederick M. Gedicks, Toward a Constitutional Jurisprudence of Reli- gious Group Rights, 1989 WIS. L. REV. 99; Michael W. Mcconnell, Accommodation of Religion, 1985 SUP. CT. REV. 1, 17-19. 51. See Mark Tushnet, The Constitution of Religion, 18 CONN.L. REV. 701, 729- 38 (1986), reprinted in MARK TUSHNET, RED, W~E,AND BLUE (1988); McConnell, supm note 50, at 19-21. 52. Freedom of association must be implied from other explicit constitutional provisions. See infm note 54 and accompanying text. Federalism may be the other explicit support for community, but that is governmental community. 53. For a critique of modern liberalism, aid by implication much modern constitutional law and interpretation, see MICHAELJ. SANDEL,LIBERALISM AND THE LIMITS OF JUSTICE(1982). For a critique of the critique, see C. Edwin Baker, Sandel on Rawls, 133 U. PA. L. REV.895 (1985). CONFLICTING PARADIGMS straining government by carving out an area free from governmental impingement. This demarcated area of freedom of religion can also be seen as part of a more general exclusion of ultimate truth from the control of the government based upon the larger structure of the Constitution as an almost exclusively procedural document and upon the First Amendment as a whole. And, finally, the restraint upon government articulated by the Free Exercise Clause supports the mediating structure of the church (or other religious groups) as a locus for activity and meaning beyond governmental interference. The second and third theoretical views are more complementary than may first appear. To the extent one sees the Constitution as a procedural document which fences government out of decisions about ultimate truth through its structure, purpose, and through the First Amendment as a whole, space for intermediate institutions is created; and, given the social nature of persons and of their intellectual and spiritual endeavors, this is a necessary space if the limit on government is to have any significant effect. It is not surprising, therefore, that the Supreme Court has implied from the First Amend- ment as a whole a right of "freedom of association" not found in the text.54 The structure and theoretical values drive toward such a conclusion. And it is significant that with the Free Exer- cise Clause there is no need for such implication; the "right of association" is there quite explicit. Before moving on, it should be noted that there is a structural view of the First Amendment and the role of the religion clauses within it that is quite different from the one sketched above. This view focuses on the content of the entire Amend- ment. Speech, press and assembly to petition for redress of grievances all involve communication, including in the latter case a gathering together for communication. Much religious activity is communicative in nature. As noted above, one key function of religion may be the determination and affirmation of ultimate truth. And prototypical religious conduct is probably worship: a gathering together for conduct which is primarily communicative. Noscitur a sociisf a maxim of statutory

54. Citizens Against Rent Control v. City of Berkeley, 454 U.S. 290 (1981); JOHNE. NOWAKET AL., CONSTITUTIONALLAW $ 16.41 (3d ed. 1986); NAACP v. Alabama a rel. Patterson, 357 U.S. 449 (1958); Roberts v. United States Jaycees, 468 U.S. 609 (1984); TRIBE, supm note 33, at $$ 12-26 & 14-16. 55. "It is known from its associates. The meaning of a word is or may be 26 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [I993 construction, suggests determining the meaning of a word from accompanying words. Religion thus interpreted with the accompanying language of the First Amendment could be read to be limited to the kinds of conduct otherwise covered by the Amendment: communication and association for the purpose of communication, with the symbolic speech conduct typical of worship being the only conduct explicitly added by reference to religion.56 Leaving the Establishment Clause aside, the First Amendment becomes a functional whole under this interpretation. In the process, however, the protection for religious conduct aside from worship disappears, and the Free Exercise Clause becomes nothing more than an explicitly mentioned instance of freedom of speech and press. Such a view is consistent with the Enlightenment-deist- rationalist perspective described above, and from this direction, a paradigm of equality for religious bases of belief with others makes some sense. As Justice Jackson articulated the position: "It was to assure religious teaching as much freedom as secular discussion, rather than to assure it greater license, that led to its separate ~tatement."~'But this perspective leaves out the radical protestant view, which may well have been domi- nantf and is difficult to reconcile with a text which is phrased as an absolute grant of freedom of religion. D. Doctrine and Precedent Thomas Jefferson appears to have preferred a view of religious freedom congruent with this last understanding of the First Amendment, a view which protects religious opinion and expression but nothing more." The operative language of his Virginia Bill for Religious Freedom was restricted to "opinion, belief, profession and argument."60 Beyond this, the government could act: "[Ilt is time enough for the rightful purposes of

known from the accompanying words." BLACK'S LAW DICTIONARY1060 (6th .ed. 1990). 56. Marshall, Free Exercise as Free Expression, supra note 6, at 545; Pepper, Alternatives, supra note 3, at 367-68. 57. Douglas v. City of Jeannette, 319 U.S. 157, 179 (1943) (Jackson, J., concurring in the result). 58. McConnell, supra note 22; Smith, supra note 37. 59. David Little, Thomas Jefferson's Rdtgious Views and Their Inf2uence on the Supreme Court's Interpretation of the First Amendment, 26 CATH. U. L. REV. 57, 58-64 (1976). 60. See supra note 19 and accompanying text. 71 CONFLICTING PARADIGMS 27 civil government for its offieers to interfere when principles break out into overt acts against peace and good order."' The operative language of the Free Exercise Clause is not limited to speech and opinion, however, and Madison, who was intimately involved in the drafting, as Jefferson was not, had a rather different view of when government could act to restrict religious conduct. His suggested language for the 1776 Virginia Declaration of Rights not only used "free exercise" as the operative language, but further stated that "no man or class of men ought on account of religion be subject to any penalties or dis- abilities, unless under color of religion the preservation of equal liberty, and the existence of the State be manifestly en- dangered."62 Jefferson's view is where the Supreme Court started with its first interpretation of the clause. One hundred years later the announced doctrine of the Court was more congruent with Madison. Then, in an abrupt turnaround in 1990, the Court returned to the Jeffersonian understanding. Reynolds v. United States" is that first Jefferson-based interpretation, and it is worth a brief detour as an example of the historical, textual, and structural concerns we have considered. The persecution of and violence aimed at the Mormons is a useful reminder of the "simple history" mentioned above and of the concern that history generated for the framers: religious dserences tend to beget violence and bloodshed. Quite literally in search of space in which to live out their religious convictions, the Mormons vacated civilization and moved to the great western desert to create their Zion in the wilderness. George Reynolds, following his religious duty as a Mormon, was married to more than one woman. Under federal law, this bigamy was a crime, but the First Amendment protected the "free exercise" of religion. From the radical protestant perspective, the polygamy of the Mormons is the garden, and the state's criminal law is an intrusion from the wilderness. If the religion clauses protect the church, the criminal law must not intrude. But from the Enlightenment perspective of protecting the state from the incursion of religion, the criminal law is a primary mechanism of government that ought to be protected from intrusive religious claims. If the Free Exercise Clause is the Enlightenment's compromise, George Reynolds presents the

61. STOKES,supm note 18. 62. Hunt, supra note 21, at 166-67. 63. 98 U.S. 145 (1878). 28 BRIGHAM YOUNG UNrVERSITY LAW REVIEW [I993 question of the bounds of that compromise: How extensive is the sheltering space? We know that Jefferson's view confines freedom of religion to opinion and belief, but George Reynolds actually married a second wife; he did not just believe in polygamy. This is conduct clearly beyond worship, or any cluster of communicative activities otherwise protected under the First Amendment. On its face, polygamy does not appear to be conduct which threatens "equal liberty" or the "existence of the State," but that may well have been arguable.64 The Court had no need to struggle with the absolute language of the clause, or with Madison's relatively absolute notions ("wholly exempt"), however, because the Justices simply relied on Jefferson for the proposition that freedom of religion protected belief only and not action. From this rather constricted initial interpretation, the Court retreated, but very slowly. In the 1940 opinion deciding Cantwell v. Connecticut, the Court announced that the Free Exercise Clause was not limited to protection of belief, it also protected "freedom to act," although to a significantly lesser degree. Freedom to believe was "absolute," but action could be regulated "for the protection of society" as long as government did not "unduly infringe the protected freedom.'s5 Cantwell was one of a long series of Jehovah's Witnesses cases in which the Supreme Court announced a great deal of First Amend- ment law, but did not clearly distinguish the parameters of free speech from freedom of religiod6 For the most part, there was little need to do so because the conduct of the Witnesses was almost exclusively communicative-classic speech, press and assembly activity such as street corner proselytization through direct conversation, the playing of records, and the

64. The opinion does rely in part on the assertion that polygamy leads to "sta- tionary despotism." Reynolds, 98 U.S. at 166. Justice Souter's concurring opinion in the Hialeah case asserts that Reynolds remains valid on this basis, meeting the Sherbert-Yoder test described below. 113 S. Ct. 2217, 2245-47 (1993). The Reynolds opinion is complex and interesting and I have analyzed it elsewhere at some length. See Pepper, Altemtives, supra note 3, at 317-26. 65. 310 U.S. 296, 303-04 (1940). Cantwell is also significant as the opinion in which the Court clarifies that the Free Exercise Clause is applicable as a limit on governmental action by the states through the Due Process Clause of the Four- teenth Amendment. The Clause has been "incorporated" into the Fourteenth Amendment so as to apply against the states. 66. See, e.g., Follett v. Town of McCormick, 321 U.S. 573 (1944); Martin v. City of Struthers, 319 U.S. 141 (1943); Murdock v. Pennsylvania, 319 U.S. 105 (1943); Chaplinsky v. New Hampshire, 315 U.S. 568 (1942); Lovell v. City of Griffin, 303 U.S. 444 (1938). See also Pepper, Altemtives, supra note 3, at 326-30. CONFLICTING PARADIGMS distribution of handbills. Free speech and press doctrine was sufficient to reach decisions concerning this kind of conduct. For this reason, and probably others as well, throughout this extensive span of precedent it was never clear that the Free Exercise Clause protected anything beyond the kinds of non- religious conduct protected by the speech, press and assembly clauses. It was thus unclear whether the Free Exercise Clause had any contemporary effect or had, instead, become only a vestigial anachroni~m.~' With Sherbert v. Vernerss in 1963 and Wisconsin v. Yoders9 in 1972 the Supreme Court clarified that the Free Ex- ercise Clause did have an ambit of its own well beyond that of other First Amendment freedoms. In doing so it articulated a doctrine and a test which seems much closer to the Madisonian vision of free exercise quoted above than to the Jeffersonian understanding relied on by the Reynolds court. In Sherbert, Adell Sherbert lost her job, and was unable to find another, because of her refusal to work on Saturday, her Sabbath. She was denied unemployment compensation benefits on the basis of having refused to accept "suitable work." In Yoder, Amish parents refused on the basis of religious belief to send their children to school for the final two years of compulsory education (ninth and tenth grades). The parents were convicted of a crime for violating the mandatory school-attendance law. In both cases the Supreme Court held that application of the law at issue to the religious conduct at issue was an impingement on freedom of religion prohibited by the Free Exercise Clause. The test announced and the language used suggested sweeping protection for religious conduct. In Sherbert the Court said that even an "incidental burden on the free exercise of appellant's religion may be justified [only] by a 'compelling state interest.' "'O Later in the opinion the Court put the matter more emphatically: "[Iln this highly sensitive constitutional

67. Under modern doctrine, governmental action intentionally or explicitly discriminating against religion is prohibited by the Establishment Clause, as is such conduct discriminating among religions. Everson v. Board of Educ., 330 U.S. 1, 15- 16 (1947); McDaniel v. Paty, 435 U.S. 618, 636-42 (1978) (Bre~an,J., concurring); Larson v. Valente, 456 U.S. 228 (1982). Including religion with race as a suspect classification under the Equal Protection Clause also has the effect of rendering most such governmental conduct unconstitutional. See supra note 1. 68. 374 U.S. 398 (1963). 69. 406 U.S. 205 (1972). 70. 374 U.S. at 403 (quoting NAACP v. Button, 371 U.S. 415, 438 (1963)). 30 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [I993 area, '[olnly the gravest abuses, endangering paramount interests, give occasion for permissible limitation.' "?' Similarly, the Yoder court stated the test as follows: "The essence of all that has been said and written on the subject is that only those interests of the highest order and those not otherwise served can overbalance legitimate claims to the free exercise of reli- gi~n."?~ Much has been written about the Court's free exercise opinions in the Jehovah's Witnesses cases and during the era when Sherbert and Yoder were the rule. This article will not attempt to cover that ground again, although it is interesting and w~rthwhile.'~Rather, the following paragraphs briefly summarize the doctrine created in the Sherbert and Yoder opinions and the remarkable inconsistency with which that doctrine was applied by the Court. The Sherbert-Yoder doctrine had five main component^.'^ First, action was protected-conduct beyond speech, press, or worship was included in the shelter of freedom of religion. Neither Sherbert's refusal to work on the Sabbath nor the Amish parents' refusal to let their children attend ninth and tenth grades can be classified as conduct protected by the other clauses of the First Amendment. Second, indirect impositions on religious conduct, such as the denial of twenty-six weeks of unemployment insurance benefits to Adell Sherbert, as well as direct restraints, such as the criminal prohibition at issue in Yoder, were pr~hibited.'~Third, as the language quoted above

71. Id. at 406 (quoting Thomas v. Collins, 323 U.S. 516, 530 (1945)). 72. 406 US. at 215. 73. See, e.g., TRIBE, supra note 33, at 1154-301; Jesse H. Choper, The Free Exercise Clause: A Structural Overview and an Appraisal of Recent Developments, 27 WM. & MARY L. REV. 943 (1986); George W. Dent, Jr., Religious Children, Secu- lar Schools, 61 S. CAL. L. REV. 863 (1988); Ira C. Lupu, Where Rights Begin: The Problem of Burdens on the Free Exercise of Religion, 102 HARV. L. REV. 933 (1989) [hereinafter Lupu, Where Rights Begin]; Ira C. Lupu, Free Exercise Exemption and Religious Institutions: The Case of Employment Discrimination, 67 B.U. L. REV. 391 (1987); Marshall, The Case Against the Compelled Exemption, supra note 6; Mar- shall, Free Exercise as Free Expression, supra note 6; Pepper, Taking, supra note 37; Pepper, Alternutives, supm note 3. The multitude of sources cited in this material will lead the reader into the literature on the subject. 74. See Pepper, Taking, supra note 37, at 308-12. 75. This is frequently called the unconstitutional conditions doctrine and stands for the proposition that a government cannot accomplish indirectly, through denial or conditioning of benefits, that which it is constitutionally prohibited from doing directly. Sherbert is a primary precedent in this line of cases which recently has become rather complex. It is a crucial doctrine for cases like Bob Jones Univ. v. United States, 461 US. 574 (1983) (non-profit status) and Quaring v. Peterson, 728 71 CONFLICTING PARADIGMS 31 indicates, the protection granted was extensive. Only extremely strong governmental interests justified impingement on religious conduct, as the absolute language of the text of the Free Exercise Clause suggests. Fourth, the strong language was backed by a requirement that the government provide proof of the important interest at stake and of the danger to that interest presented by the religious conduct at issue. Fifth, in determining the injury to the government's interest, a court was required to focus on the effect that exempting religious claimants fkom the regulation would have, rather than on the value of the regulation in general. Thus, injury to governmental interest had to be measured at the margin: assuming the law still applied to all others, what would be the effect of exempting the religious claimant in this case and other similarly situated religious claimants in the future?76Together, the fourth and fifth elements required that facts, rather than speculation, had to be presented concerning how the government's interests would be harmed by excepting religious conduct from the law being challenged. Sherbert and Yoder adopted a balancing test for free exercise jurisprudence, and balancing tests are notoriously manipu- lable.77Exaggeration of the weight on the governmental interest side of the balance is particularly likely through speculation about the effects of decisions adverse to those interests and through defining those interests at a higher level of generality than the constitutional interests on the other side of the balance. For example, in Prince v. Massachu~etts~~an aunt took her niece proselytizing on the peaceful evening streets of Brockton. Sarah Prince, the aunt, was convicted of violating child labor laws. Despite the fad that none of the evils ordi-

F.2d 1121 (8th Cir. 1984), afd by equally divided Court sub nom. Jensen v. Quaring, 472 U.S. 478 (1985) (per curium, Powell, J., did not participate) (driver's license), where what is at issue is in the form of a government "benefit." In Sherbert the Court stated: "[Tlo condition the availability of benefits upon this appellant's willingness to violate a cardinal principle of her religious faith effedive- ly penalizes the free exercise of her constitutional liberties." 374 U.S. at 406. 76. This fifth element is simply an application of the more general principle frequently referred to as the Yeast drastic means" analysis. This is usually presented as an inquiry into whether the government can reach its goal or serve its interest through means which do not impinge-or which impinge significantly less--on the constitutional right at issue. In this f&h element of the Sherbert- Yoder doctrine, the "less drastic means" is exempting religious claimants from an otherwise valid legal regulation. 77. See Pepper, Alternutives, supra note 3, at 341-44, and sources cited there. 78. 321 U.S. 158 (1944). . 32 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [I993 narily associated with child labor were present in the case, the Supreme Court placed on the state interest side of the balance "the interest of youth itself, and of the whole community, that children be both safeguarded from abuses and given opportuni- ties for growth" and the "crippling effects of child employment, more especially in public place^."'^ The Court did not similarly inflate the other side of the balance. The effect of the fourth and fifth elements of the Sherbert- Yoder doctrine was to impose a discipline which helps prevent this kind of manipulation. In Sherbert the Court would not accept, without proof, speculation about the effect of possible future fraudulent religious claims on the compensation fund.80 The government's interest in compulsory education is clearly "compelling," and ranks "at the very apex" of state f~nctions.~' But in Yoder the Court refused to put this general interest on the balance, and instead focused upon the state's marginal interest in compulsory education of Amish children after the eighth grade, a matter of quite different weight. Exempting Amish children from required attendance for the ninth and tenth grades has relatively little effect on the state's generally compelling interest in public education." The Sherbert-Yoder doctrine was not applied consistently by the Supreme Court. Mrs. Quaring's case appeared to be an easy one under the doctrine because exempting the very few persons who would object to having photos on their driver's licenses would seem likely to have very little, if any, deleteri- ous effects on the interests of the state. (New York did not require photos on licenses until just a few years before the Quaring case, and Nebraska itself had several exceptions such as learner's permits.) Nonetheless, the United States Supreme Court split four to four, and because opinions are not issued in cases of tie votes, no explanation was given.83 The situation presented in Bob Jones University was far more diffi~ult.~~The university engaged in race discrimination mandated by religious belief. Because of this conduct, the Inter-

79. Id. at 168 (citations omitted). 80. 374 U.S. 398, 407 (1963). 81. Yoder, 406 U.S. at 213. 82. Id. at 221-29. 83. See Quaring v. Peterson, 728 F.2d 1121 (8th Cir. 1984), afd by equally diviokd Court sub nom. Jensen v. Quaring, 472 U.S. 478 (1985) (per curiam, Powell, J., did not participate). 84. Bob Jones Univ. v. United States, 461 US. 574 (1983). 71 CONFLICTING PARADIGMS 33 nal Revenue Service revoked its tax exempt status, a major blow to a contemporary non-profit educational institution. The case was problematic because the effect of an exemption from race discrimination prohibitions in education based upon religious convictions is hard to gauge. There is some evidence that fraudulent claims to such beliefs might be the result in some areas given the resistance to integrated education. But unlike the Yoder opinion where the Court carefully analyzed and weighed the interests on both sides, in Bob Jones it dealt with the constitutional question in three brief paragraphs. The Court cited both Sherbert and Yoder as the appropriate prece- dents (among others), and repeated the kind of sweeping protection enunciated in those cases: in order to "justify a limitation on religious liberty," the state must show "that it is essential to accomplish an overriding governmental intere~t."'~It proceeded to recognize the "compelling" nature of the government's general interest "in eradicating racial discrimination in education."86 The fourth and fifth elements of the Sherbert-Yoder doctrine then disappeared, however, as the Court simply asserted that there were "no less restrictive means . . . available to achieve the governmental interest," but did not discuss why exemption was not an acceptable less drastic means in Bob Jones when it was in Yoder, or what effect such an exemption would have on the governmental intere~t.'~ In Unites States v. Lee,s8 a case dealing with the Amish mandate to live "separate and apart" in regard to social security taxes rather than education, the Court recited the tests from Sherbert and Yoder, and then seemed not to apply them in reaching the conclusion that the religious conduct was not protected. Instead of narrowing the government interest, the Court inflated it in a way reminiscent of the Prince de~ision.~' In the case of Bowen v. Roy," however, sincere religious objection to obtaining a social security number was held protected. Even though a social security number was a statutory requirement for receiving federal welfare benefits, exemption from

85. Id. at 603 (quoting United States v. Lee, 455 U.S. 252, 257-58 (1982)). 86. Id. 87. Id. at 604 (citations omitted). See Douglas Laycock, Tax Exemptions for Racially Discriminatory Schools, 60 Tnr. L. REV. 259 (1982). 88. 455 U.S. 252 (1982). 89. Id. at 259-62. For a brief discussion of Lee, see Pepper, Conundrum, supra note 3, at 299-302. 90. 476 U.S. 693 (1986). 34 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [I993 that requirement was mandated by the C~urt.~'Justice O'Connor wrote a ringing endorsement of the Sherbert-Yoder doctrine in a separate opinion in Roy in 1986," but in 1988 she wrote an opinion for the Court finding the doctrine entirely inapplicable to sincere religious objections by American Indians to planned construction of an access road through national forest land which included their most sacred sites.93 In sum, the Court articulated the Sherbert-Yoder doctrine with some consistency, but only rarely applied it. Then, in a case in which the validity of the doctrine was neither ques- tioned nor argued, the Court abruptly and surprisingly denied that it had ever existed, and appears to have returned to a reading of the clause similar to that used in Reynolds. Justice Scalia authored the opinion for a five member majority. Assert- ing that the Court had "never held that an individual's religious beliefs excused him from compliance with an otherwise valid law prohibiting conduct that the State is free to regulate," the opinion held that usage of peyote as a genuine, central religious practice of the Native American Church was not sheltered from criminal prosecution by the Free Exercise Cla~se.'~

91. There is some disagreement as to the holding in Bowen resulting from five opinions. A close reading of the opinions reveals that five justices followed the Sherbert-Yoder doctrine finding it unconstitutional to require the father to provide a social security number for his daughter as a condition for his receiving welfare benefits. See Pepper, Taking, supra note 37, at 319-22. This understanding of the opinions in the case appears to be conhed in Hobbie v. Unemployment Appeals Comm'n, 480 U.S. 136, 140-42 (1987); but see the concurring opinions of Justices Powell and Stevens, id. at 146-48. 92. 476 U.S. at 724 (O'Co~or,J., concurring in part and dissenting in part). 93. See Lyng v. Northwest Indian Cemetery Protective Ass'n, 485 U.S. 439 (1988), discussed in Lupu, Where Rights Begin, supra note 73. See also Pepper, Conundrum, supm note 3, at 281, for a brief discussion of the determinative issue in Lyng. Justice O'Connor also wrote an elaborate dissent in Goldman v. Weinberger, 475 U.S. 503, 528 (1986), a case dealing with the uniform requirements of the military conflicting with the wearing of a yarmulke by an orthodox Jew. The Court held against the religious claimant in that case and in O'Lone v. Estate of Shabazz, 482 U.S. 342 (1987), a case dealing with a free exercise claim by a prison inmate. In both cases, I would suggest that the holdings result more directly from a wariness of the Court concerning the relation between the Bill of Rights and total institutions in general, than from a rejection of the Sherbert-Yoder doctrine. See Pepper, Taking, supra note 37, at 322-23, and cases cited supra note 109. 94. Employment Div. v. Smith, 494 U.S. 872, 878-79 (1990). Justice O'Co~or authored a concurrence arguing that the compelling interest test (what I have called the Sherbert-Yoder doctrine) was clearly the ruling and accepted interpretation of the Free Exercise Clause, but concluding that the government had met its burden of showing a compelling interest. Id. at 891. Justice Blackmun, with whom Justices Bre~anand Marshall joined, dissented, agreeing with Justice O'Comor 71 CONFLICTING PARADIGMS 35 In so ruling, the Court reached out to decide an issue not presented. The facts dealt with unemployment compensation, not criminal prosecution. And the Oregon Supreme Court had ruled that the criminality of the conduct was (1) irrelevant to the award or denial of unemployment compensation in Oregon and (2) hypothetical and hence irrelevant in the absence of a criminal pr~ceeding.'~Justice Scalia's opinion for the Court was remarkable for several reasons: (1) its lack of candor regarding the prior twenty-five years of precedent under the Sherbert- Yoder doctrine, (2) its refusal to honor the Oregon Supreme Court's interpretation of its own law concerning the irrelevance of uncharged criminal conduct to the award of unemployment compensation, and (3) its reaching out to undo a doctrine not put at issue in the briefs or oral argument and to decide an issue not genuinely before it. The opinion has been subject to well-deserved, sharp and extensive criticism, and efforts have been made to map out what content the opinion may have left for the Free Exercise Clause.96 I shall not attempt to tread through that thicket here. It should be noted, however, that there are several factors which suggest that the doctrine announced in Smith might be short-lived. First, Justice O'Connor, who agreed with the result in Smith but authored a sharp and powerful concurrence asserting the continued vitality and justice of the Sherbert-Yoder doctrine, has emerged over the last two terms as one of the leaders of the new, independent minded, centrist block on the Court, including herself and Justices Kennedy and Souter. Second, Justice Souter, who was appointed to the Court after the Smith decision to replace Justice Brennan, wrote a separate concurrence in the recent Hialeah decision expressing his disagreement with and criticizing the Smith ruling, indicating his understanding that the Sherbert-Yoder doctrine remains good law in conflict with Smith, and suggesting reasons why the doctrine of stare decisis does not preclude reconsideration of the Smith rule." This was particularly significant because

that the compelling interest test was the correct doctrine, but finding that, under the fads, the government had not shown such an interest. Id. at 907. 95. See Smith v. Employment Div. 11, 763 P.2d 146, 147 & n.3, 148 (Or. 1988); Smith v. Employment Div. I, 721 P.2d 445 (Or. 1986). 96. Excellent analysis of the case and what may be left of free exercise doctrine can be found in Laycock, supra note 5, and McCo~ell,supra note 5. 97. Church of the Lukurni Babalu Aye v. City of Hialeah, 113 S. Ct. 2217, 2241-50 (1993). 36 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [I993

Justice Souter agreed with the unanimous result in Hialeah and was not on the Court which considered Smith: the only purpose of the concurrence was to put on the record his disagreement with a recent prior ruling, a rather unusual act for a relatively new Justice, and one clearly intended to signal lawyers and lower court judges that Smith might not be reli- able precedent. Third, Justice White, one of the five votes supporting Justice Scalia's revision of the meaning of the Free Exercise Clause, has resigned from the Court. Ruth Bader Ginsburg, nominated to succeed him, is likely to be more inclined toward a liberty-oriented understanding of religious freedom, although there appears to be little indication on the record of her inclinations in this regard.g8(On the other hand, Justice Kennedy, the third member of the influential central group, joined the Smith decision and authored the Hialeah majority decision which relied explicitly upon Smith. This would make it more dacult for him to change his position and join Justices O'Connor and Souter in a return to something like the Sherbert-Yoder doctrine. And Justice Thomas, who replaced Justice Marshall, joined in all but one portion of Justice Ke~edy'sHialeah decision.ggThus, the Court appears to re-

98. Judge Ginsburg dissented from the denial of a rehearing en banc in the free exercise case of Goldman v. Secretary of Defense, 739 F.2d 657 (D.C. Cir. 1984), writing a brief opinion supportive of freedom of religion where the issue was the wearing of a yarmulke contrary to military regulations concerning the uniform. (Interestingly, then Judge Scalia, the author of Smith, joined her opinion.) The case was later decided by the Supreme Court against Doctor Goldman, the free exercise claimant. Goldman v. Weinberger, 475 U.S. 503 (1986). The Supreme Court's determination is best understood as part of a long line of decisions limiting first amendment freedoms in the special, limited rights contexts of the military and prisons. See Pepper, Taking, supra note 37 at 322-23. Judge Ginsburg has also authored two opinions applying the Sherbert-Yoder doctrine; one ruling in favor of a free exercise claimant; the other ruling against the claimant, finding a compelling governmental interest to support application of the law at issue. Leahy v. District of Columbia, 833 F.2d 1046 (D.C. Cir. 1987) (overturning a summary judgment for the District, returning the case to the trial court, noting that "the District has not demonstrated that requiring a religious objector to provide his social security number in order to obtain a driver's license is the least restrictive means of achieving the concededly vital public safety objec- tive at stake," id. at 1049, see notes 2, 83, 144, and accompanying text); Olsen v. Drug Enforcement Administration, 878 F.2d 1458 (D.C. Cir. 1989) (upholding denial of free exercise protection for the use of marijuana; the fads of the case provide an interesting comparison with Smith, supra note 94). Judge Ginsburg also joined in the opinion in Wilson v. Block, 708 F.2d 735 0.C. Cir. 1983), a controversy concerning Native American objection to development of federal land, analogous to the Lyng case, supra note 93. 99. 113 S. Ct. at 2221. 71 CONFLICTING PARADIGMS 37 main in ap uneasy five to four division on whether Smith or Sherbert-Yoder is the correct understanding of the Free Exer- cise Clause.loO)Fourth and finally, on May 24, 1993, the House of Representatives passed the Religious Freedom Resto- ration Act of 1993, which would reinstate the "compelling state interest" standard and is intended to undo the Smith decision. There appears to be substantial support in the Senate as well.lol Thus, the doctrinal interpretation of the Free Exercise Clause began by adopting Jefferson's Enlightenment-oriented view and ignoring the quite different view of Madison and the radical protestants. This was an understanding compatible with an equality based paradigm. Over the fifty-year period beginning with Cantwell in 1940, however, the Court developed (but did not consistently follow) a well articulated doctrine aligned more with the Madisonian and radical protestant understanding-a doctrine clearly reflecting an understanding of the clause as mapping out an area of liberty and freedom from government impingement. With Smith, a five member majority turned away from the liberty paradigm-seemingly, although not clearly, turning back to an equal treatment mod- ello2-and left free exercise doctrine in an undeveloped, unclear state. And now, with the resignation of one of those five, and with Justice Souter's declaration of disagreement, future interpretation of the clause is both open to substantial change and difficult to predict.

100. Justices Scalia, Ke~edy,Thomas, Stevens and Chief Justice Rehnquist support the Smith rule; Justices O'Connor, Souter, and Blackmun (and possibly future Justice Ginsburg) support the Sherbert-Yoder doctrine. 101. For a discussion of the Act, see Douglas Laycock's article in this symposium. Douglas Laycock, The Religious Freedom Restoration Act, 1993 B.Y.U. L. REV. 221. 102. This term's opinion in the Hialeah case confirmed this understanding. Justice Kennedy's opinion for the Court relied upon Smith for "the general proposition that a law that is neutral and of general applicability need not be justiiied by a compelling governmental interest even if the law has the incidental effect of burdening a particular religious practice," and overturned the Hialeah laws because they were neither formally neutral nor generally applicable. 113 S. Ct. at 2222. And Justice Blackmun in concurrence stated Smith was "wrongly decided, because it ignored the value of religious freedom as an affirmative individual liberty and treated the Free Exercise Clause as no more than an antidiscrimination principle." Id. at 2250. 38 BRIGHAM YOUNG UNrVERSITY LAW REVIEW 11993

E. Values, Policy and Prudence: The "Why" of the Free Exercise Clause Why religion? The First Amendment singles out religion for both a special benefit-the guaranty of "free exerci~e'~-and a special detriment-the prohibition on "establishment." No other aspect of life is similarly singled out in the Constitution, not science, education, art, philosophy, family relations, or agriculture. Commentators have begun to ask about the Free Exercise Clause: Why should there be a special freedom for religious conduct?103When I first heard that question put at an academic conference on the religion clauses, I thought the answer, at least in constitutional terms, was obvious. But perhaps it is not nearly so obvious as I had thought. How else can one explain the contrast between the absolute language of the clause and the Supreme Court's stingy interpretation? How else can one explain cases like Quaring and Smith? There is a strong current of equality in American values, and equality is particularly identified with the Constitution. It thus goes against our grain to grant religious believers legal preference (such as exemption from valid legal regulations) that others do not receive. And so the grain is often followed despite a constitutional pointer in the other direction which could hardly be clearer. Why religion? The answer, to a large and meaningful extent, is obvious. That answer is: text and history. Both, I submit, are remarkably clear. And when those primary sources of constitutional interpretation are clear, that alone is an important reason. The "simple history" discussed briefly above provides a rather succinct explanation of why the framers saw religion as sufficiently different to carve out two explicit constitutional provisions concerning it. Their history taught them that religion meant trouble, serious blood-spilling trouble, when it was mixed with government. Their purpose was to try to ensure civil peace in regard to religious questions. Their solution was to try to separate government and religion, and the recognition that the separation had to go both ways: religion could not interfere with the mech-

103. See, e.g., John H. Garvey, Free Exercise and the Values of Religious Liberty, 18 CONN. L. REV. 779 (1986); Michael E. Smith, !Z%e Special Place of Religion in the Constitution, 1983 SUP. CT. REV. 83. 71 CONFLICTING PARADIGMS 39 anisms of the state, but the mechanisms of the state must not interfere with the garden of the church. From Hobbes and Locke through Jefferson and Madison there was lasting concern and thought, on the Enlightenment side, about achieving civil peace in regard to religion.lOl And this was mirrored, al- beit to a lesser extent, on the radical protestant side with concern that religion not become enmeshed in the profane business of the state. The church must leave the state alone, and the state must leave the church alone. (The "why" in historical terms is clear; it is the "how" that is difficult. The directive in the First Amendment is clear enough; it is carrying it out which is so difficult.'" Religion and state cannot be kept separate for each deals with everyday life in the real world, as the cases from Reynolds through Quaring and Bob Jones demonstrate. Keeping the civil peace in light of the explosiveness of religious beliefs is a clear reason, and reason enough, for the religion clauses. That the elabora- tion of the directive through precedent and doctrine is difficult is not a warrant to ignore the clear message of the text and its historical context.) The nature of religion explains both this explosiveness and the fact that it does not fit comfortably within the processes of a democratic polity. Religious beliefs tend to be (1) strongly held; (2) of special significance to the believer, often relating to his or her basic identity and understanding of both self and the reality beyond self; and (3) beyond verification or testing by rational discourse or other decision making or knowledge gathering processes such as voting or scientific investigation. Major- ity rule does not fit well with those kinds of beliefs; compromise does not fit a great deal better; and imposition by legal authority from above on such beliefs is not likely to be taken with equanimity. Religion is differentieven singular (what else compares?)lo6-and that singularity is perfectly congruent with both the text and history of the Free Exercise Clause.

104. ROGERS M. SMITH, LIBERALISM AND AMERICAN CONSTITUTIONALLAW 19 (1985). But cf. Smith, supra note 37. 105. The practical difficulties, although asserted summarily more than seriously examined, were a major part of Justice Scalia's justifcation for the abandonment of the Sherbert-Yoder doctrine in the Smith opinion. See 494 U.S. at 882-90. 106. John Garvey compares it to insanity. Garvey, supra note 103, at 798. Al- though I think that comparison is not particularly apt, it is suggestive of the dominant secular view of our age and of how far we have travelled from the 1787 understandings. 40 BRIGHAM YOUNG ~IVERSITYLAW REVIEW [I993

That singularity is also consistent with a constitutional decision to treat religion differently, to give it substantial shelter from decisions reached through majoritarian democratic processes, substantial shelter from otherwise valid law. (Given the singular nature of religion, the clarity of the text, and the coherence of the historical background, what may be difficult to understand is the disinclination to grant the shelter. The answer is the strong pull of equality mentioned above, the difficulty in framing a doctrine with workable limits [no freedom can be absolute], and other factors mentioned below.) The discussion so far in this section explains the special place of religion in the polity for the most part on the basis of the threat religion poses to civil peace, a negative cast attribut- able primarily to the Enlightenment view. But the singularity of religion contains a promise-a positive cast-as well which also supports its place in the First Amendment and generous exemptions for believers. That promise, for at least some believers, is a life lived in a dimension beyond the ordinary, a life with connections and obligations different in kind from the physical and social world we all share; a promise of life lived in accordance with transcendent truth and with a profound connection to that truth. Those who have an intimation of that promise, who live at least part of their lives with such connections, tend to give it an importance, a valuation, an allegiance, above all others. It tends to give a meaning to life that those of us without profound connection to that dimension do not share. That meaning, that connection, is often perceived, by both believer and unbeliever, as good. It is perceived as making life better. If she had a choice (and sometimes choice is involved) the believer would choose that life over one without the transcendent connection. And many who do not believe would also choose that life if they could. The threat and the promise arise from the same phenomenon: religious experience. The strength and profundity and higher allegiance of religion threaten the peace of the state; but they also promise a life of a different and better kind for the believer. Why would the state want to impinge on that life if it could avoid it? The believer suffers a terrible choice; the state puts itself at risk. What is to be gained by putting the believer to a test of allegiances? There are two simple answers. First, the state gains whatever the object is of the law or regulation which is impinging on the believer: all the ubiquitous purposes of common action 71 CONFLICTING PARADIGMS 41 through government. Is that gain worth the loss and the risk? I submit that it is usually not. The Free Exercise Clause seems to be a determination on the constitutional level that it is not. The Sherbert-Yoder doctrine, with its identification of "overriding" and "compelling" governmental interests as the necessary ground for limiting free exercise and with its search for 'less drastic means" (which do not impinge on believers) to reach those interests, appeared to rest on that same conclusion. Moreover, under that doctrine most of the purposes of governmental action could still be reached: application of the Free Exercise Clause created an exemption for the believer; the law remained valid for all others,lO' and thus remained generally effective. lo' Second, the state gains equality in the application of its law, a central value, a constitutional value. But the factors discussed above suggest that religion is different in a significant way: the person whose religious life is invaded by a legal provision is not similarly situated to the person for whom the provision has no such effect. The impact of the legal provision on those differently situated persons is not equal. Also, the risk to the state in enforcing the law is not equal. Thus there is good reason for unequal application of the law, and it is a constitutionally sufficient reason under the Equal Protection Clause. (This leaves aside for the moment the fact that this particular inequality is based upon the explicit text of the Con- stitution-the Free Exercise Clause.) There is another value to religion, another reason for treat- ing it differently. The added dimension of a religious life includes not only a profound connection to the truth, but also characteristically a profound connection to the community de- fmed by allegiance to that truth.logThus, part of the promise of religion is transcendent meaning in the communal life as well as the individual life. Many find in our current law, including constitutional law, an overemphasis on individualism

107. Ira C. Lupu, Keeping the Faith: Religion, Equality and Speech in the U.S. Constitution, 18 CONN.L. REV. 739, 761-78 (1986). 108. If the exemption would be applicable to a substantial minority, it is unlikely the legal provision would have been enaded to begin with. See Pepper, Taking, supra note 37, at 313-15; McCo~eu,supra note 5, at 1147-48. Thus the exemption doctrine is unlikely to substantially undermine the effectiveness of government. 109. See, e.g., Note, Reinterpreting the Religion Clauses: Constitutional Construc- tion and Conceptions of the Self, 97 HARv. L. REV. 1468, 1472-74; Gedicks, supra note 50, at 106-15 and other sources cited supm note 50. 42 BRIGHAM YOUNG UNIVEXSITY LAW REWIEW [I993 and autonomy and a corresponding insufficient recognition of and support for community and co~ection.'~~Given the well recognized communal dimension in most religions, the Free Exercise Clause is the preeminent exception: the value of corn- rnunity and connection are given explicit constitutional protection. Only here is communal activity (other than governmental activity) given constitutional shelter."' The more one is concerned with creating legal and constitutional support for corn- munal life, the more one will be inclined toward an interpretation which gives the Free Exercise Clause power and vitality. '12 F. The Synergism of Text, History, Structure and Policy Focusing on the speech and press provisions of the First Amendment, and in doing so finding protection of communicative expression to be the central concept, William Marshall believes that the equality paradigm provides the proper interpretation for the Free Exercise Clause. Under this view, only expressive religious conduct is protected from governmental imposition by the Clause, and only to the same extent it would be under the speech and press clauses if it had no religious content or m~tivation."~Professor Marshall has elaborated . arguments against the understanding of each of the factors presented above, and each of these arguments has some weight. But the arguments have a distinct ad hoc sense about them, in that they attack each basis with whatever appears at hand-in the nature of a legal brief-rather than cohering

110. See, e.g., MARYANN GLENDON,RIGHTS TALK: RIE IMPOVERISHMENT OF PO- LITICAL DISCOURSE(1991). 111. Federalism, the existence of the states, is the other protection for community in the Constitution. But few would think the states now, or at the time of the adoption of the Constitution, sufficiently small to support the values associated with community life. 112. Religion is not so singular in regard to its communal dimension as it is in the ways discussed above, and thus the equality problem is more salient. Other areas of life-families for example-share a fundamental communal dimension. But religion is a place to begin, and it has a textual basis in the Constitution. Interestingly, the Supreme Court has felt the need to create constitutional shelter for aspects of family life even without a textual basis. See, e.g,, Zablocki v. Redhail, 434 U.S. 374 (1978) (right to marry); Moore v. City of East Cleveland, 431 U.S. 494 (1977) (right to live with extended family); Loving V. Virginia, 388 U.S. 1 (1967) (right to marry); Pierce v. Society of Sisters, 268 U.S. 510 (1925) (liberty to direct upbringing of children); Meyer v. Nebraska, 262 U.S. 390 (1923) (right to pursue a legitimate vocation); TRIBE, supra note 33, at chs. 15-20. 113. Marshall, The Case Against the CompeUed Exemption, supra note 6. CONFLICTING PARADIGMS together.ll4 I do not intend to address each of these arguments or attempt to rebut each. What I want to point out here is that such an approach neglects the way in which the factors developed above interact with and support one another. In a nutshell, Professor Marshall's argument is that text and history do not make it clear enough that the Free Exercise Clause was intended to extend to protection from otherwise valid, religiously neutral laws. When it comes to the area of values/policy/prudence, however, his argument is that religion is not sufficiently unique; other non-religious bases for action sometimes share some of the aspects that justify special constitutional treatment for religion. What this view neglects is that religion is unique in the way it combines the values/policy/prudence reasons with an explicit textual basis and substantial, concordant historical and structural reasons. The text of the First Amendment includes what appears to be a clear and near absolute declaration of a zone of liberty for religious "exercise." That plain reading seems extreme and difficult to reconcile with notions of government in general and equality in particular, however, until the structural logic of such a position is explained. Further, when examined in the light of both the long- and short-term history preceding the drafting and adoption of the religion clauses, both the text and the structural logic which support it make a great deal of sense. In particular, the perspective of the radical protestants, their political importance in the evolution of the separation of church and state in the colonies, and the articulated views of James Madison, tend to give a distinct coherence to this combination of text, structure and history.l15 What Professor Marshall's perspective and criticism leave out is this syner-

114. For example, the argument concerning the. fact that the Free Exercise Clause is rendered close to content-less by an equality based interpretation is an- swered by pointing out that the Clause is left with some imaginable meaning; and that subsequent passage and development of the Equal Protection Clause can hardly determine the original or current meaning of the Free Exercise Clause. Mar- shall, The Case Against the Compelled Exemption, supra note 6, at 373-74. Marshall's argument from history is, essentially, that it is too unclear to be determinative. Id. at 375-79. (McConnell, supra note 22, has provided substantial historical evidence to support exemption doctrine, but Marshall still has a significant point.) Note that the textual and historical arguments do not join or create mo- mentum for one another; and this is what I mean by ad hoc. The sense is of an effort to find something wrong or weak with each argument, whatever that might be. 115. This is similar to what Professor Steven Smith calls the "religious justifica- tion" for religious freedom. See Smith, supra note 37, at 154-66. 44 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [I993 gism: those aspects of the factors which support a liberty paradigm, which support exemptions for religious conduct from neutral law, interact with and support one another in a way which makes the whole substantially stronger than the parts. Professor Marshall notes, for example, that it is not only religion which gives some persons strong and sometimes ada- mant bases for action. In addressing the kind of argument presented in part I1.E. above, he emphasizes that religion is not "uniq~e.""~(While it is true that other bases of belief occasiondly can result in similar depth and emotion, it is probably also true that religion as a category is unique: such strength and depth and resistance to rational compromise are associated with religion with far greater frequency than with any other basis for action. Thus religion is in the aggregate unique, even if there are numerous cases of similar results in the absence of a religious basis.) In addition, he points out that many religious believers do not hold their beliefs with the strength idunwillingness to compromise which makes religion arguably unique, and therefore exemptions for all sincere religious believers from neutral laws would be "overbroad." That is, some who would get the benefit of the exemptions would not have the depth or kind of belief which justifies the exemption."' But this aspect of Professor Marshall's argument leaves out one factor of primary significance: the text. The First Amendment expressly protects religious conduct; it does not expressly protect any other bases for conduct.l18

116. Marshall, The Case Against the Compelled Exemption, supra note 6, at 383 (providing the example of one unavailable for work not because of Sabbath obligations or religious objections to work in an armaments factory, Thomas v. Review Board, 450 U.S. 707 (1981) (following Sherbert u. Verner, 374 U.S. 398 (1963)), but "due to strong convictions about parental obligations") (footnote omitted). 117. Marshall, The Case Against the Compelled Exemption, supra note 6, at 384. This problem is addressed, in part, in Pepper, Taking, supra note 37, at 327-28. In the end, Professor Marshall admits that while "no one fador" is conclusive, "the aggregation of a number of fadorsn may mean that religion is unique and therefore "entitled to special protection." Marshall, The Case Against the Compelled Exemption, supra note 6, at 385-86 (commenting on Garvey, supra note 103). Professor Marshall appears recently to have come to a firmer conclusion that religion is unique, or at least sufficiently different to require special constitutional and political treatment. See William P. Marshall, The Other Side of Religion, 44 HASTINGSL.J. 843 (1993) [hereinafter Marshall, The Other Side of Religion]. His observations and conclusions there appear quite consistent with the discussion of religion in part 1I.E. 118. It does expressly protect other, inherently limited, categories of conduct: speech, press, and assembly for redress of grievances. See supra the beginning section of part 1I.A. and the concluding two paragraphs of part 1I.C. 71 CONFLICTING PARADIGMS 45 That the reasons for this special protection may not apply to each instance of religious conduct, and that other conduct may occasionally share the special nature of religious motivation, would not seem to be of great import in light of the express singling out of religion in the text of the Amendment. (We are not dealing here with a large and conceptually vague provision such as "equal protection" or "due process.") Thus Professor Marshall's argument leaves out the connection, the mutual support, between the policy and values basis and the explicit textual support for a special liberty paradigm singling out religion for special treatment. Once he has left the subjects of text and history, Professor Marshall's argument tends to forget them. For example, at the conclusion of one argument11g he states: "Again, those advo- cating a free exercise exemption for religious groups must con- vincingly argue that religious exercise is special."'" Striking- ly, the text of the First Amendment, which unequivocally singles out religious "exercise" for unique treatment, is not referred to and appears to have been forgotten, or to have become somehow insignificant. The values and policy part of Professor Marshall's argument thus reads as if the First Amendment had not yet been written.121 As such, if presented in a textual and

119. Responding to an equality aspect of the liberty paradigm, to be developed briefly infra at part 1V.A. 120. Marshall, The Case Against the Compelled Exemption, supra note 6, at 380. 121. Marshall, The Case Against the Compelled Exemption, supra note 6, at 379- 86. In this section he quotes Professor Mark Tushnet, another participant in this symposium: "In a pluralistic society with crosscutting group memberships, the over- all distribution of benefits and burdens is likely to be reasonably fair." Id. at 380 (quoting Mark Tushnet, The Emerging Principle of Accommodation of Religion (Du- bitante), 76 GEO.L.J. 1691, 1700 (1988)). This is clearly written as if there were no Bill of Rights, and asserts no need for one. Interestingly, it is similar to one of the reasons given by Madison for the proposition that the union of the states would have a tendency to break and control the violence of fadion . . . . [Tlhe greater number of citizens and extent of territory . . . renders factious combina- tions less to be dreaded . . . . Extend the sphere, and you take in a greater variety of parties and interests; you make it less probable that a majority of the whole will have a common motive to invade the rights of other citizens; or if such a common motive exists, it will be more mcult for all who feel it to discover their own strength and to ad in unison with each other. THE FEDERALISTN0.10 (James Madison). Whether or not one agrees with this argument in the context of the communica- tions and politics of the contemporary United States, or with Tushnet's similar reason, both were clearly rejeded by the adoption of the Bill of Rights as an explicit protection from majority rule in regard to those matters covered by the first 46 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [I993 historical vacuum-that is if we had no Bill of Rights and were considering what to include in one-it might well present a persuasive argument that we ought to confine religious freedom to an ambit identical with freedom of speech only. But we are not in such a vacuum. The combination and interaction of the textual, historical, structural and values/policy/prudence foun- dations for a liberty-based interpretation, which grants shelter to religious conduct from formally neutral law, is substantially more powerful than it appears from the perspective of criticism which considers each basis separately. And in doing constitutional law we seek one interpretation of one set of words which will function as a guiding rule or principlethe basis for the development of coherent doctrine. We are not looking for and separately evaluating a textual, an historical, a structural, and a policy Constitution. We use all these perspectives as interpretive tools, as alternative angles of vision, to assist in a unified understanding of a particular legal provision.122

When the Constitution and First Amendment were drafted, the rational Enlightenment approach to the world was relatively new and still struggling with the theretofore dominant religious understanding. The Constitution is itself a landmark in the movement toward the dominance of a secular, rational view of living in the world. In the ensuing two hundred years, secularization has dramatically increased and become clearly dominant in Western intellectual and legal thought.lB With this ten amendments. Indeed, adoption of such a list of freedoms was an explicit re- quest by several states attendant to (and an implicit condition for) their vote to ratify. See MILLER,supra note 21, at 119-21. In an October 17, 1788 letter to Jefferson, Madison claimed that he had "always been in favor of a bill of rights; provided it be so framed as not to imply powers not meant to be included in the enumeration." He expressed concern, however, that "the most essential rights could not be obtained in the requisite lati- tude," specifying in that regard "the rights of conscience in particular." 5 THE WRITINGSOF JAMESMADISON 269-75 (This is the famous letter in which Madison goes on to discuss the "inefficacy of a bill of rights," to refer to them as mere "parchment barriers," and to discuss the fad that under the Constitution, as opposed to previous governments, "the danger of oppression lies in the interested majorities of the people rather than in usurped ads of Government"). 122. See supra sources cited at note 8. 123. That secularization has become dominant in these arenas does not mean that it is necessarily dominant in society generally or in the lives of most individuals, although it may be. Professor Steven Smith provides substantial sociological 71 CONFLICTING PARADIGMS 47 paradigm shift, approaching the religion clauses from the En- lightenment view becomes close to automatic: the protection of government from the pernicious influence of religion is easy to grasp, whereas the special protection for religious conduct looks like unjustifiable discrimination. The Establishment Clause remains coherent, but it becomes difficult to give content to the Free Exercise Clause. The threat of religion is perceived more easily than its promise, and religion becomes marginal. Similarly, with the transition to the liberal paradigm for most legal thought, the focus on the individual as the locus of rights and legal status makes it harder to perceive intermediate groups or community as independent bearers of legal rights and status. From this direction also the Enlightenment view of the Establishment Clause as a primary protection for the government (and indirectly for individuals) comes to predominate over a view concerned with and protective of religious communities-the latter a view more consistent with the republican tradition. We are thus far removed from the framers' context in which strong constitutional protection for religion, including religious communities, made intuitive sense. It is no great surprise, therefore, that until Sherbert and Yoder the Free Exercise Clause was reduced to the other First Amendment freedoms, in effect given no independent content.'* For these same reasons, it is also unsurprising that the Court found it diflkult to consistently apply the doctrine announced in those cases, and has now returned to a view which provides religious freedom little or no special protection. Contemporary interpretation and application of the Free Exercise Clause thus faces a choice. One can read the clauses from Jefferson's perspective, an Enlightenment view that sees religion primarily as threat. Or one can turn to Madison (a different Enlightenment view) and to the radical protestants, data suggesting that religion is a live force in most American lives. He appears to grant the dominance of the secular view in political, intellectual, and legal thought, but prefers to refer to this as the privatization of religion rather than the secular- . ization of these arenas. Smith, supra note 37, at 169-80. Otherwise, the view of the Free Exercise Clause presented there and that argued here are, for the most part, consistent and compatible. 124. Tushnet, supra note 51, at 729-38. I remain uncomfortable with the ambi- guity of the usage "liberal" in this context, but will employ it nonetheless. In the more generic usage of riberal," I am not sure it can be identified more with an individualistic than a comrnunitarian approach to law. 125. See supra text accompanying notes 56-65. 48 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [I993 and see religion as both a threat and a promise.126 The Jeffer- sonian view leads to a strong Establishment Clause, and to a Free Exercise Clause with no contemporary effect. Religious conduct is protected if it falls within the protections of the speech, press, or association provisions of the First Amend- ment, but not otherwise.12' The Madisonian and radical protestant views lead to a strong Free Exercise Clause while not denigrating from a strong Establishment Clause. The absolute text of the Free Exercise Clause is hard to reconcile with the restrictive view, as is much of the historical context. An understanding of the Free Exercise Clause under which it has vigor and effect, on the other hand, fits well with both text and historical context. It also makes sense from several perspectives on the structure and theory of the Constitution in general and the First Amendment in particular. The values, policies and prudence suggested by the text, its history, and a significant part of the theory and structure of the Constit~tion'~~also point in this direction. Choosing to take the text seriously, choosing to interpret the Free Exercise Clause as granting substantial protection for religious conduct from governmental regulation, is only a first step. Such an orientation turns religion clause jurisprudence down a road which requires a great deal of doctrinal development. A freedom of such large scope cannot be absolute, yet the clause is phrased without limit. Neither history nor structure provides clear limit or gloss. The construction of limits is a daunting task and may explain, at least in part, the Supreme Court's delay in taking the initial step and its apparent inconsistency in application of the Sherbert-Yoder doctrine. And it is the primary reason presented in the Smith opinion for undoing that d0~trine.l~~ Charting the contours that a vigorous but workable free

126. See supra parts IIA. & B. 127. See Marshall, The Case Against the Compelled Exemption, supra note 6; Marshall, Free Exercise as Free Expression, supra note 6. 128. It cannot be denied that si-cant theoretical and structural aspects of the Constitution support the Jeffersonian approach. As mentioned supra, text accompanying note 56, the First Amendment can be read as a unit protecting fundamentally communicative activity only (including religious speech, press and possibly worship to the extent it includes symbolic speech, but no other kinds of religious activity). Also, as mentioned in part II.E., the emphasis of our constitutional tradition on equality does not square well, on Grst thought, with a constitutional protection for religious conduct that extends to no other kind of conduct. 129. 494 U.S. at 882-90. 71 CONFLICTING PARADIGMS 49 exercise doctrine might take is a necessary task for constitutional interpretation. It is a larger task than I am willing to undertake in this article.1s0 To give a sense of the scope and viability of the task, however, a few of the more prominent difficulties that must be faced in the application of a robust interpretation of the Free Exercise Clause are discussed briefly below.

IV. A QUARTETOF DIFFICULTIES A. The Establishment Clause, Equality, and Exemptions fiom Neutral Law The presence of the Establishment Clause complicates free exercise doctrine, and the Free Exercise Clause likewise complicates Establishment Clause doctrine. As noted above, the First Amendment singles religion out for both a special detriment and a special benefit. If the Establishment Clause prohibits all legal preferences for religion, the Free Exercise Clause is itself a violation. Likewise, if the Free Exercise Clause prohibits all government detriments to religion, the Establishment Clause is itself a ~iolation.'~~In an age of ubiquitous governmental regulation and subsidy, this has become a serious com- plication for First Amendment analysis. Either clause can be interpreted so expansively as to leave little content for the other. And whatever interpretation is given one clause, a compatible interpretation can be given the other. There are a number of ways of approaching this tension between the clauses. Any resolution must also face the connect- ed question of what space will be left between the clauses for discretionary governmental action neither compelled by the Free Exercise Clause nor prohibited by the Establishment Clause.ls2 These approaches will not be canvassed here. I will pause, however, for a brief focus on exemptions and establishment. Granting believers exemptions from neutral legal regula-

130. Some of my earlier work has made some suggestions in this area. See, e.g., Pepper, Taking, supm note 37, at 325-36; Pepper, Conundrum, supra note 3. See also J. Morris Clark, Guidelines for the Free Exercise Clause, 83 HARV.L. REV.327 (1969); Marc Galanter, Religious Freedom in the United States: A Turning Point? 1966 WIS.L. REV. 217. 131. Jesse H. Choper, The Religion Clauses of the First Amendment: Reconciling the Conflict, 41 U. Prm. L. REV. 673 (1980); Pepper, Altemtives, supm note 3, at 345-52. 132. McConnell, supra note 50. 50 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [I993 tions that apply to others is a benefit for religion based upon an explicitly religious criterion, and thus can be seen as a violation of the Establishment Clause. (This is the route one takes to find an equality paradigm to be the appropriate reading of the Free Exercise Clause.) But there is nothing to distinguish that conclusion from one which reads the entire Free Exercise Clause out of the Constitution, for that Clause also is a benefit for religion based upon an explicit textual religious criterion: "free exercise of religion." There is a structural understanding of the clauses, however, with which exemptions are perfectly congruent; and this understanding is perfectly congruent with the Free Exercise Clause as a guaranty of substantive rather than merely formal equality. Assume for the moment that the absolute language of the Free Exercise Clause is premised on the Madisonian notion, sketched above,'= that there is no consent to be governed in regard to matters of religion, and that the clause is the "social contract" to that effect incorporated into the Constitution. This is perfectly in tune with the understanding of the Bill of Rights as a charter protecting the minority from the majority. When combined with an effective Establishment Clause, minority and majority will be in the same position in regard to the nexus between religious practice and government action. The clauses together prevent the majority from purposefully using government to favor itself or disfavor minorities in religious matters (in other words, separation of church and state). But the majority is also unlikely to unintentionally impinge on its own religious beliefs (with which it is quite familiar), whereas it is quite likely to do so in the case of small religious minorities. Consider for a moment, would states require photos on driver's licenses if Jewish or Catholic religious beliefs prohibited- use of photographs?ls4 To truly equalize minority and majority (or a coalition of minorities) in regard to the relations between religion and government, the clauses must therefore protect against not only intentional discrimination on religious matters, but also the inadvertent. Thus the need for, and constitutional logic of, exemptions for believers from neutral governmental actions

133. See supra part 1I.C. 134. See Pepper, Conundrum, supra note 3, at 296-99 & 11.159; Pepper, Taking, supm note 37, at 312-16; Laycock, supra note 5, at 14-15; McConnell, supm note 5; at 1147-49. 71 CONFLICTING PARADIGMS 51 that remain valid as to others.135And this constitutional logic and structure leave the Establishment Clause intact and with a strong independent role of its own. In fact it leaves essentially all of the Court's current Establishment Clause doctrine intact. ls6 The inadvertence of the government's religious discrimination, and the equalization of majority and minority, lead us to the question of perspective: the government's conduct is inno- cent and nondiscriminatory in plan and intent; but in effect it is oppressive to the minority. If we are interpreting a document, such as the Bill of Rights, designed to protect minority from majority, the question of which perspective is appropriate would seem easy to answer. Moreover, the text, history, structure and values implicated in interpretation of the clauses, discussed above, all point toward this perspective as well. Thus substantive equality-a reading of the religion clauses which leaves both politically dominant and politically weak religious groups equal in their inability to use the government (law) to assist their own religion or burden others-makes the most sense as an interpretation of the Bill of Rights, a document designed to protect minorities and individuals from democracy (the majority or a coalition of minorities).ls7 one other, related equality concern ought to be mentioned here. Seeing in the First Amendment primarily a protection for speech, Professor Marshall suggests that a liberty paradigm for religion gives it an improper advantage in the marketplace of ideas that is inconsistent with his understanding of the free

135. Galanter, supra note 130, at 291; Pepper, Taking, supra note 37, at 312-15; Comment, A Non-Conflict Approach to the First Amendment Religion Clauses, 131 U. PA. L. REV. 1175, 1197-208 (1983). 136. It should probably be mentioned, as well, that the view articulated in this paper which leads to a vigorous, powerful Free Exercise Clause can also lead to a similarly effective Establishment Clause. Personally, while I believe Establishment Clause doctrine leaves a great deal to be desired, P am not convinced that very many of the Supreme Court's cases have been wrongly decided. And most of those that I believe are wrong, such as Lynch v. Donnelly, 465 US. 668 (1984) (the nativity scene case), and Mueller v. Allen, 463 U.S. 388 (1982) (tax deductions for parents for some parochial school expenses), err on the side of failing to find an establishment violation where one has occurred. 137. Justice Souter's significant concurring opinion this term in the Hialeah case, see text accompanying note 97, identifies the Smith decision with formal neutrality and the Sherbert-Yoder doctrine with substantive neutrality, bringing that terminology into the free exercise opinions of the Court for the first time, and arguing the latter as the correct interpretation of the free exercise clause. 113 S. Ct. at 2240- 42. 52 BRIGHAMYOUNGUNIVERSITYLAWREVIEW [I993 speech provisions of the Amendment.ls8 From this perspective, if protection for religious communication is greater than for other communication, religion is given an advantage over other bases for conduct in the free competition of ideas. This difficulty is easily remedied, however, if the First Amendment is read as providing religious communication (speech and press) the same degree of protection as other core areas of protected expression. To the extent religious conduct is only communication (speech or press) otherwise covered by the Amendment, there appears to be an overlap of protection, and no need or reason to provide extra protection for religious speech over political speech. Only when the religious conduct is not primarily communicative (and thus would not be protected under the speech and press clauses) is there need or reason for a Free Exercise Clause. (And thus, of course, the significance of a clause that uses a word-"exercise"-which so clearly denotes conduct of all kinds, conduct beyond the scope of communication, conduct which is neither speech nor press.) Polygamy, ingestion of peyote, refusal to carry a picture or to send one's children to school, all may have communicative or symbolic elements; but they are not primarily speech, and would not fall under the current interpretations of the scope of the speech and press clauses of the First Amendment. When the conduct is primarily communicative, however, there is no need to discriminate between religious and non-religious conduct. In Widmar u. Vincent,13' the Supreme Court held that because the University of Missouri had created an open forum for student speech on campus, it could not discriminate against religious speech (interpreting a worship service by students to be speech). In doing so, it held that the Free Speech Clause (and equal treatment under the open forum doctrine), rather than the Establishment, Free Exercise, or Equal Protection Clauses, was the appropriate standard. Such a view directly meets Professor Marshall's concerns in the area of speech, and is the prevailing view on the Co~rt."~(And even if it were not, it must be remembered that there is no Establishment

138. Marshall, The Case Against the Compelled Exemption, supra note 6, at 388- 94. 139. 454 U.S. 263 (1981). 140. In addition to Widmar, see Heffkon v. International Soc'y for Krishna Con- sciousness, Inc., 452 U.S. 640 (1981); cf. Texas Monthly, Inc. v. Bullock, 489 U.S. 1 (1989). The Court has left this somewhat less clear than it was before through dicta in the Smith case. See Laycock, supra note 5, at 44-47. 71 CONFLICTING PARADIGMS 53 Clause for speech. Thus government is not prohibited from favoritism in regard to ideas and bases for conduct in its own speech. Public schools are free to teach that capitalism is better than communism, respect and concern for the environment better than despoliation, and they commonly do so. There is no requirement that the government be neutral or treat all ideas equally in its own c~mmunication.)'~~ A First Amendment doctrine which treats religious communication equally with other communication is not enough from Professor Marshall's view, however, because the Free Exercise Clause can still be read to give preferential treatment to religious conduct beyond speech, and this violates his understanding of the First Amendment in its entirety as an "equality of ideas" pro~ision.'~~There is some validity to this concern, some real unfairness in giving advantage to those who act from religious motivation. (Although, as pointed out at the beginning of this article, that unfairness varies greatly depending upon the context in which the claim arises.) But it must be noted that this is not an unfairness in giving a preference where the Constitution is concerned with the marketplace of ideas: the speech and press clauses. And it must be emphasized that there is no "equality of ideas" clause in the Bill of Rights, while there is an explicit Free Exercise Clause. Moreover, the government favors some ideas and disfavors others in all its vast criminal and civil regulation of conduct. It criminalizes theft; it encour- ages charitable contributions (and defines what counts as charitable).143When Professor Marshall moves his argument from speech to conduct, it conflicts directly with the ubiquitous governmental regulation of conduct in this society. In respect to religion, government merely does so at a more fundamental level of legal preference, by including freedom of religion in its basic charter of rights. B. Drawing Lines, Neutrality and Discrimination Applying the Free Exercise Clause under the Sherbert- Yoder doctrine requires that lines be drawn and discrimina- tions be made. The sincere must be distinguished from the

141. See generally MARK G. YUDOF,WHEN GOVERNMENTSPEAKS (1983). 142. Marshall, me Case Against the Compelled Exemption, supra note 6, at 392- 94. This view resonates with the structural unity of the First Amendment articulated in the text accompanying note 49. 143. Cf. Wisconsin v. Mitchell, 113 S. Ct. 2194 (1993). 54 BRIGHAM YOUNG UNIVERSIW LAW REVIEW [I993 fraudulent. (For example, someone who objects to mandatory education for non-religious reasons may be tempted to claim the religious exemption granted the Amish.) Instances where religious claimants cannot be exempted from government regulations because of the extent of damage resulting to important government concerns must be distinguished from situations where the exemption inflicts tolerable damage to the governmental interest. (For example, a person who asserted religious objection to carrying any driver's license or alternative official identification would present a difficult judgment even under a vigorous free exercise interpretation; a situation far more problematic than presented by Mrs. Quaring.'") When such dis- tinctions are drawn there is a lack of equality not just between those with a religious basis for exemption and those without, but also amongst those asserting religious claims. Some justices of the Supreme Court have been unwilling to adopt a doctrine under which such lines must be drawn, perceiving that the basic command of the religion clauses is neutrality: that government not discriminate between religions. But if the Free Exercise Clause is to have modern content, such lines must be drawn. If sincerity cannot be considered, or if no interest of the government justifies impingement on religion, we are close to an absolute freedom. Little if any content will be given to such an unlimited freedom. Thus, on the claim of avoiding the danger of the kinds of religious discrimination the clauses were designed to prevent, these justices end up granting essentially no meaningful protection for religious conduct other than against overt religiously discriminatory government action.145 To protect those claimants who might lose because

144. Automobiles are dangerous instrumentalities; unrestricted use by unidentifi- able persons may create too much risk to tolerate, too complete a preference for religion, particularly in light of the non-consenting third parties who may be harmed. On the other hand, it might be that "less drastic means" could be created by a sufficiently flexible government even for this situation. Those who had a genuine religious need not to carry a license (sincerity would be an important dimension of such a case) might register with the state, provide identifying information, and sign a waiver granting, in any situation in which a driver's license is normally required by law enforcement personnel, permission for detention until identification could otherwise be confirmed. This would certainly be a signiFicant administrative burden on the state, but because the number of people claiming the exemption is likely to be very small, the occasions for detention would likely be extremely rare. (Such a waiver would have the incidental effect of helping to test the sincerity of the religious claim.) 145. For a more extended discussion of the sincerity issue and of the positions of Justices White and Stevens, see Pepper, Taking, supra note 37, at 325-31. CONFLICTING PARADIGMS

they are incorrectly found to be insincere, or the government interest is found too important, all religious exemptions must fall. To protect the feelings of that subgroup of religious claimants, all persons whose religious conduct is harmed by neutral government action must suffer.'" (One is reminded of de- stroying the village to save it.) This is simply another facet of the Enlightenment-oriented approach which, seeing religion primarily as threat rather than promise, elevates the Establishment Clause protection of the government from religion over the Free Exercise Clause protection of religion from g~vernment.'~'Meaningful contemporary religious liberty under the Free Exercise Clause entails difficult choices in the course of application; and this is certainly a significant cost. Sometimes such line drawing will be painful, sometimes it will seem unjust, and sometimes it will be errone- ous. The alternative neutrality view may avoid these costs, but it must be remembered that it does so at the probably greater cost of reading out of the Constitution one of its most explicit and fundamental rights. '48

146. Douglas Laycock characterizes this view as a preference for "even-handed repression" over "imperfect liberty." Laycock, supra note 5, at 14. 147. The Establishment Clause retains vigor under the neutrality view. Govern- ment cannot discriminate among religions in the granting of assistance or benefits, and it cannot discriminate in favor of religion generally by granting it benefits or assistance other similar activities or institutions do not receive. Of course this creates one major discrimination: the constitutional provision discriminating against religion retains vigor and effect; the provision discriminating in favor of religion does not. Professor Kurland has been the foremost academic proponent of the "neutrality" approach, proposing that the clauses be read as a limitation "akin to . . . the Equal Protection Clause," prohibiting use of religious classifications "for purposes of governmental action, whether that action be the conferring of rights or privileges or the imposition of duties or obligations." PHILIP B. KURLAND,RELIGION AND THE LAW 17-18 (1962) discussed in Pepper, Alternatives, supra note 3, at 346-48. 148. It is far too easy to forget, or not to notice, the simple fact that the "neutrality" approach is hardly neutral or equal in impact. Laws which impinge on religious conscience inflict an injury of a quite different kind on the believer than the inconvenience or general constraint on liberty that same law inflicts on one whose religion is not affeded. Thus neutrality and equality may lie in the eye of the beholder, and may depend on whether one looks at the face of a legal provision or at the effect on the claimant. For a discussion of similar issues concerning the importance of perspective, see Martha Minow, Foreword: Justice Engendered, 101 HARV. L. REV. 10 (1987). Justice Souter makes a similar point in his concurrence this term in the Hialeah case. There he states: "Our cases make clear, to look at this from a different perspective, that an exemption for sacramental wine use would not deprive Prohibition of neutrality. Rather, 'such an accommodation [would] 'reflect nothing more than the governmental obligation of neutrality in the face of religious differ- 56 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [I993

C. Administrability and Bureaucracy Judicial opinions are opaque, particularly to non-lawyers. Case law decides a dispute retrospectively based upon a comparison of the facts in the disputed situation with the facts in the precedent cases. It applies by analogy; it is law reached by deciding which facts were the relevant ones leading to the decision in the prior opinion, because rarely are the facts in a subsequent situation substantially the same as those in the precedent case. Such law is dissimilar from legislation or regulation because those are usually formulated as discrete rules; rules which, on their face at least, are more precise and delimited. Free exercise jurisprudence is exemplary of this problem. The Free Exercise Clause is not a precise or delimited rule which on its face can guide the action of governmental actors. The Sherbert-Yoder doctrine was complex and subtle, requiring simultaneous application of a number of factors and percep- tions; it exhibited and demanded flexibility in its application. And any doctrine sacient to create a vigorous but workable body of free exercise law will be similar, even if it is different from Sherbert-Yoder in significant ways. Judicial opinions applying and creating free exercise law are unlikely to be simple or easy to understand. This creates a real danger that the law as applied on a day-to-day basis by non-lawyer government actors will be quite different from the law as it would be decided by the courts. Bureaucracies, through which most government activities occur, are particularly unlikely to be flexible, subtle or complex in their understanding and application of free exercise doctrine. Uniformity and rules-lack of discretion-are their preferred mode of operation. Overgeneralization from a judicial decision, or series of decisions, is thus a real possibility. A contemporary example from Establishment Clause doctrine is instructive: the recent cleansing from public school texts of references to and discussions of religious subjects149appears to have occurred, at least in signi.fkant part, because of a misunderstanding and overextension of Establishment Clause requirements by school boards, school administrators, text book committees and text ences.' ' " 113 S. Ct. at 2241 n.2 (quoting Yoder, 406 U.S. at 235 n.22, quoting Sherbert, 374 U.S. at 409). 149. Paul C. Vitz, Religion and Traditional Values in Public School Textbooks, 84 PUB. INTEREST 79 (1986); Dent, supra note 73. CONFLICTING PARADIGMS

book publishers.150 This happened despite repeated and explicit dicta by the Supreme Court that it was not required by their decisions-repeated statements that studying religion or religious material in the context of a secular subject like history or literature was acceptable and that it was religious practice (such as prayer) or indoctrination that was unconstitution- a1.151 The extent of this problem varies greatly with different factual contexts. For example, the bureaucrats who administer driver's licenses are likely to be able to differentiate between someone in Mrs. Quaring's position and the claimant hypothe- sized above whose religious beliefs prohibit use of any driver's license at all or any formal substitute. And in the Bob Jones situation, IRS agents are more likely to have the training (and the skepticism) to apply a pro free exercise decision in a flexible and subtle way than are other bureaucrat^.'^^ Given the subtleties and charged feelings in the school situation, school officials, as the Establishment Clause example suggests, may be more likely to misunderstand and overgeneralize. In Mozert u. Hawkins County Public Schools,153funda- mentalist parents claimed that imposition of a particular basic reading series in public school was a violation of their free

150. The causes of this phenomenon are probably complex, and would be an important subject for empirical research. Mixed with misunderstanding of Supreme Court doctrine may well be (1) a preference to avoid even coming near the constitutional line, (2) a preference for an excuse to avoid the sensitive and controversial questions of religion in public education, and (3) a bias that religion is appropriate in the private realm and inappropriate in the public realm. This, of course, would be in accord with the Enlightenment view of religion and with the dominant modern secular intellectual paradigm. See Dent, supra note 73, at 867-73; Pepper, Takings, supra note 37, at 306-07. For an argument that religion is inappropriate in the public realm, see Marshall, The Other Side of Religion, supra note 117, which contains references to some of the current writing on that question. 151. See, e.g., Epperson v. Arkansas, 393 U.S. 97, 106 (1968); Abington Township Sch. Dist. v. Schempp, 374 U.S. 203, 225 (1963); McCollum v. Board of Educ., 333 US. 203, 235-36 (1948) (Jackson, J., concurring). 152. This may be over-optimistic given the IRS' initially slow and ineffective response to the Universal Life Church claims of tax exemption for their "churches." See Universal Life Church v. United States, 372 F. Supp. 770 (E.D. Cal. 1974); Note, Mail Order Ministries, the Religious Purpose Exemption, and the Constitution, 33 TAX LAW. 959 (1979-80); Stephen Schwartz, Limiting Religious Tax Exemptions: When Should the Church Render unto Caesar, 29 FLA. L. REV. 50, 62 m.81 & 82 (1976); see also Heins, "Other People's Faiths": The Scientology Litigation and the Justiciability of Religious Fraud, 9 HASTINGSCONST. L.Q. 153 (1981). 153. 647 F. Supp. 1194 (E.D. Tenn. 1986), rev'd, 827 F.2d 1058 (6th Cir. 1987). The facts, the two opinions, and the basic issues the case raised are explored in detail in Stolzenberg, supra note 49, at 581. 58 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [I993 exercise rights because of the content of many of the readings. The federal district court in a careful and nuanced application of the Sherbert-Yoder doctrine agreed, and ordered that the students be exempted from reading class and allowed to receive home instruction in reading. The judge was well aware of the potential for overgeneralization: While the Court must be sensitive to the widespread implications of its decisions, it must also limit its decisions to the facts of the case before it . . . . This opinion shall not be interpreted to require the school system to make this option available to any other person or to these plaintiffs for any other subject. Further accom- modations, if they must be made, will have to be made on a case-by-case basis by the teachers, school administrators, Board, and Department of Education . . . .I5' The court's attempt at care notwithstanding, this would seem to be exactly the kind of widely publicized decision likely to be misunderstood and overgeneralized by administrators, bureaucrats, elected public officials (such as school board members) and parents, most of whom are far more likely to hear of and be influenced by the decision than to have read the opinion. The defendants and many commentators in the media were concerned that such a decision would cause crippling administrative problems for many public schools. As written it probably would not have; but as misunderstood by a large segment of relevant actors, as it might well have been, it could have led to significant problems of over-application. This is a genuine concern for courts elaborating free exercise doctrine, and may explain some of the inconsistency in application of the Sherbert-Yoder doctrine.ls5 It is the kind of concern that might underlie a preference for the neutrality approach discussed in the section above, for avoiding distinc- tions in application makes the doctrine much simpler. Sherbert was the law for over twenty-five years, however, and I know of

154. 647 F. Supp. at 1202-03. 155. In United States v. Lee, 455 U.S. 252 (1982), for example, the Court's apparent failure to follow the Sherbert-Yoder doctrine may have stemmed from concern that an opinion dealing with the narrow situation of the Amish and social security tax might be misunderstood as applying to taxes in general, including income taxes. The opinion may also be based on the related concern that religion based exemptions from taxation are more likely to create sincerity and fraud problems than are exemptions from other governmental regulations. For a discussion of Lee, see Pepper, Conundrum, supm note 3, at 299-302. 71 CONFLICTING PARADIGMS 59 no serious problems that have been caused by overgeneralization in app1i~ation.l~~It is possible, of course, that such effects have occurred, because I also know of no studies of the effect of modern free exercise doctrine on decisions outside the courts. Because a doctrine providing for robust exemptions under the Free Exercise Clause goes against the grain of the dominant secular intellectual paradigm, however, overgeneralization may be much less of a threat here than it was with application of the Establishment Clause to the public schools, where the doctrine tended to go with the grain rather than against it.15'

D. Timidity, Limits and Possibility Even during the Sherbert-Yoder era, the decisions of the Supreme Court in favor of religious exemptions concerned, for the most part, matters not of major significance from the government's point of view. A few more persons qualifying for unemployment insurance,'" or accommodating a few welfare recipients who refuse to use a social security number,'" are hardly issues of general imp~rtance.'~~Yoder, of course, is the clear exception given the importance of education. But even in that case the consequences were distinctly circumscribed by the insular, limited nature of the Amish community and way of life. And the Court went out of its way-to an almost bizarre extentto clearly indicate that the holding was limited to the circumstances of the Ami~h.'~'Had the case been likely to have more generalized consequences for mandatory public education systems, it is unlikely it would have been decided as it

156. At least in the courts, quite the contrary was the case. Plaintiffs asserting freedom of religion claims tended to lose. James E. Ryan, Note, Smith and the Religious Freedom Restoration Act: An Iconoclastic Assessment, 78 VA. L. REV. 1407 (1992); see also Pepper, Conundrum, supra note 3. 157. See supra note 150. 158. Sherbert v. Verner, 374 U.S. 398 (1963); Frazee v. Illinois Dep't of Employ- ment Sec., 489 US. 829 (1989); Hobbie v. Unemployment Compensation Appeals Comm'n, 480 US. 136 (1987); Thomas v. Indiana Employment Sec. Div., 450 U.S. 707 (1981). 159. Bowen v. Roy, 476 U.S. 693 (1986). 160. That they are not issues of major consequence does not mean that they are not of great importance to the individuals involved, or that how the state treats such individuals is not itself of major importance. Here I am addressing the consequences, from the state's point of view, of accommodating these individuals in the particular contexts in which the Court has held it was rewired by the Free Exer- cise Clause. 161. Pepper, Alternatives, supra note 3, at 333-45. 60 BRIGHAM YOUNG UNIVERSITY LAW REVIEW 11993 was.l'j2 Thus, even under the Sherbert-Yoder doctrine, protection for religion has been held required only in areas of marginal concern to the state. The lower courts have ranged significantly further than the Supreme Court, as the Mozert trial court opinion shows, but timidity is present in those courts as well, as exemplified by the circuit court's overturning of that very decision? In this area there seems to be a distinct disinclination on the part of the courts to constrain government in any way that has major consequences. The timidity may be due in part to concern that the freedom has no limits (as the language in the Clause has no limits), a concern that giving the Clause real power may start the law down a slippery slope. I have suggested elsewhere several limits to constrain the freedom within manageable bounds, limits which would make incursions into significant governmental interests quite modest, yet yield substantial areas of freedom from government imposition for religiously motivated c~nduct.'~A coherent, limited doctrine which gives real power to the Free Exercise Clause can be constructed without undue difficulty. It will not be easy to apply-painful and problematic choices will have to be made-but it can be created with relatively minor additions to and modifications of the basic ideas underlying Sherbert and Yoder. Instead of focusing on relatively narrow limits, however, consider for a moment what would happen if the courts were less timid, if the limits were more capacious. If Bob Jones Uni- versity had maintained its tax exemption, would a great deal in our society change? There might be some incentive for fraudulent claims of religious belief to give shelter for racial discrimination in education, but both the IRS and our courts are sufficiently competent at sifting truth from falsehood that no large social change or damage is likely to ensue. More difficult is the Reynolds case. Could our society tolerate-make room for-a group with a significantly different understanding of the nature of something so fundamental as marriage? What would the consequences have been of islands of polygamy within a general rule of law mandating monogamy? Would the legal require-

162. Pepper, Altemtiues, supra note 3, at 344. 163. Mozert v. Hawkins County Pub. Sch., 827 F.2d 1058 (6th Cir. 1987). Sever- al lower court decisions are examined in Pepper, Conundrum, supm note 3, at 265. 164. See Pepper, Taking, supra note 37, at 325-35 CONFLICTING PARADIGMS ment of monogamy tend to change because of the demands of equality? Would the social fabric change in any important way? Would the non-legal forces toward conformity have led the Mormon church to repudiate polygamy in any case? What would be the consequences of a constitutional rule that shelters some fundamentally different ways of living? This stretches the imagination. The direction it moves one's thought is interesting: perhaps a society more genuinely diverse than ours; perhaps a tolerance level of a different order; perhaps pluralism in things that matter; perhaps space created in the interstices of the bureaucratic state for communities and communal values to come into being and grow.lB5And perhaps there are rather darker possibilities to imagine as well.lB6 To return to the question of equality: would it be fair, assuming such a leavening could occur in our legal regime, for the space to be available only to those with religious motivation, only to those who form religious communities? The interpretation and arguments elaborated above suggest that the answer is clearly "yes." If you are inclined toward a %o" answer, however, at least two thoughts should be considered. First, the good of such an open, diverse polity may be worth the cost of that unfairness. And, second, this strong equality incli- nation (which is what the perception of unfairness must be based upon) may have a sort of hydraulic effect: granting significant liberty for religion might over time initiate a dynamic leading to the creation of other areas of constitutional liberty. The definition of religion might enlarge over time to create such liberty;16' or other parts of the Constitution might be

165. In this regard it is interesting to note that in the last decade two of the "Forewords" to the Harvard Law Review's annual Supreme Court issue have been based on Free Exercise Clause cases, but have been concerned with far broader matters. See Cover, supra note 41; Frank I. Michelman, Foreword: Traces of Self- Government, 100 HARV. L. REV. 4 (1986). Each seems to have been imagining a legal regime with more room for difference, with significant legal space for community. Imagining a Free Exercise Clause taken seriously leads in this direction. 166. See Stephen Pepper, The Case of the Human Sacrifice, 23 ARIZ. L. REV. 897 (1981), for some imagining along these lines testing the outer limits of freedom of religion. See also Stanley Hauerwas, Self-Sacrifwe as Demonic: A Theological Response to Jonestown, in VIOLENCEAND RELIGIOUSCOMMITMENT: IMPLICATIONS OF JIMJONES'S PEOPLE'S TEMPLE MOVEMENT (Ken Levi ed., 1982), for a discussion of the Jim Jones tragedy. The recent debacle involving the Branch Davidians in Waco, Texas, is another reminder of the violent possibilities involved in religious freedom. 167. William Marshall argues that my earlier expression of such a possibility means that he and I essentially agree on the equality view. Marshall, The Case 62 BRIGHAM YOUNG UNIVERSITY LAW REVIEW 11993 their source, as has occurred with the substantive due process rights of "privacy."168

Two hundred years ago religion was recognized as a legitimate source of non-governmental authority; it was a rival to the state. The religion clauses recognize that fact. Now, quite far removed from that recognition, do we reduce religion to all other human motivations, generally subordinate to law and government? Or do we take the Constitution's guaranty of religious freedom seriously, interpreting the First Amendment to shelter significant areas of conduct, perhaps hoping that such protection might over time generalize to other related, but non-religious, motivations? Do we interpret the Constitution to expand or contract individual liberty and intermediate community power? Text, history, structure, prudence and precedent all support a vigorous understanding of freedom of religion. There are risks and difficulties with the interpretation of the Free Exercise Clause suggested here, but the promise appears greater than the threat.

Against the Compelled Exemption, supra note 6, at 402-03, commenting on Pepper, Taking, supra note 37, at 331-32. I believe this is incorrect. In the absence of the development of an expanded definition of religion, it is clear to me that the Free Exercise Clause ought to be interpreted to shelter religious conduct as traditionally understood. This includes the peyote usage in Smith, the polygamy in ReywZds (depending upon the outcome of something like the Sherbert-Yoder "compelling interest" test), and the parents' claim in Yoder, situations where I believe Marshall would not now find the Free Exercise Clause providing shelter. For me, the expansion of the definition of religion is quite secondary to the issue of reading the Free Exercise Clause as shelter from otherwise valid law. For Marshall, the definitional issue would be primary, the expansion would have to occur ,before he would read the Clause as providing protection for conduct beyond speech. See Marshall, Free Exercise as Free Expression, supra note 6, at 587-88 & n.214. For discussions of defining religion under the First Amendment see Jesse H. Choper, Defining "Reli- gion" in the First Amendment, 1982 U. IU. L. REV. 579; Kent Greenawalt, Religion as a Concept in Constitutional Law, 72 CAL. L. REV. 753 (1984); Pepper, AZtem- tives, supra note 3, at 353-64, and sources cited in those articles. 168. See TRIBE, supra note 33, at ch. 15.